WEBVTT 00:00:00.000 --> 00:00:00.580 align:middle line:90% 00:00:00.580 --> 00:00:02.440 align:middle line:84% The following content is provided 00:00:02.440 --> 00:00:06.340 align:middle line:84% by MIT OpenCourseWare under a Creative Commons license. 00:00:06.340 --> 00:00:08.500 align:middle line:84% Additional information about our license 00:00:08.500 --> 00:00:10.870 align:middle line:84% and MIT OpenCourseWare in general 00:00:10.870 --> 00:00:12.730 align:middle line:90% is available at ocw.mit.edu. 00:00:12.730 --> 00:00:16.156 align:middle line:90% 00:00:16.156 --> 00:00:18.530 align:middle line:84% GEORGE CHURCH: 101 or [? EBIO101, ?] 00:00:18.530 --> 00:00:21.110 align:middle line:90% all the other numbers. 00:00:21.110 --> 00:00:23.360 align:middle line:84% Usually we'll start these with a very brief, 00:00:23.360 --> 00:00:29.250 align:middle line:84% one slide overview of last lecture 00:00:29.250 --> 00:00:32.340 align:middle line:84% with a slightly different angle on some of the topics, 00:00:32.340 --> 00:00:33.060 align:middle line:90% and then-- 00:00:33.060 --> 00:00:36.510 align:middle line:84% and then we'll go over a preview of today's. 00:00:36.510 --> 00:00:41.900 align:middle line:84% So last session, was more an emphasis 00:00:41.900 --> 00:00:44.360 align:middle line:84% on the biological side of computational biology, 00:00:44.360 --> 00:00:46.648 align:middle line:84% and this one will be more computational side 00:00:46.648 --> 00:00:47.690 align:middle line:90% of computational biology. 00:00:47.690 --> 00:00:49.610 align:middle line:84% But obviously, both of them are-- 00:00:49.610 --> 00:00:52.640 align:middle line:84% we're trying to interweave them already. 00:00:52.640 --> 00:00:55.670 align:middle line:84% In the first one, we were specifically 00:00:55.670 --> 00:00:58.280 align:middle line:84% looking at the simplest components and the simplest 00:00:58.280 --> 00:01:03.980 align:middle line:84% systems in living systems and computational systems, 00:01:03.980 --> 00:01:06.740 align:middle line:90% and how self assembly-- 00:01:06.740 --> 00:01:10.670 align:middle line:84% here defined broadly to include symbiotic relationships 00:01:10.670 --> 00:01:17.630 align:middle line:84% between different living forms and different human inventions. 00:01:17.630 --> 00:01:22.310 align:middle line:84% And self assembly is a very critical part of biology. 00:01:22.310 --> 00:01:28.340 align:middle line:84% In the mathematics that we use, both symbolic and numeric, 00:01:28.340 --> 00:01:31.040 align:middle line:84% you must know about the approximations 00:01:31.040 --> 00:01:33.920 align:middle line:84% when we talk about the various theories going 00:01:33.920 --> 00:01:40.730 align:middle line:84% from the somewhat subatomic to the population level. 00:01:40.730 --> 00:01:44.000 align:middle line:84% There's an approximation at each level that subsumes 00:01:44.000 --> 00:01:46.220 align:middle line:90% previous approximations in it. 00:01:46.220 --> 00:01:51.390 align:middle line:84% And when you represent floating point numbers and computers, 00:01:51.390 --> 00:01:52.860 align:middle line:84% there are further approximations, 00:01:52.860 --> 00:01:56.300 align:middle line:84% which can accumulate as you do calculations. 00:01:56.300 --> 00:01:57.950 align:middle line:84% And we saw some of the ways of coping 00:01:57.950 --> 00:02:01.400 align:middle line:84% with this, such as using higher precision 00:02:01.400 --> 00:02:05.480 align:middle line:90% arithmetic in Mathematica. 00:02:05.480 --> 00:02:09.679 align:middle line:84% The other aspect of this was differential equations 00:02:09.679 --> 00:02:12.140 align:middle line:84% as a tool in studying replication, 00:02:12.140 --> 00:02:14.660 align:middle line:84% in particular autocatalytic systems 00:02:14.660 --> 00:02:19.770 align:middle line:84% which were illustrated by this really 00:02:19.770 --> 00:02:22.600 align:middle line:84% simplest possible differential equation where 00:02:22.600 --> 00:02:28.410 align:middle line:84% the incremental increase in y is a function of time, 00:02:28.410 --> 00:02:34.710 align:middle line:84% the vertical and horizontal axis is a direct function of y. 00:02:34.710 --> 00:02:37.230 align:middle line:84% A simple linear function of the-- 00:02:37.230 --> 00:02:41.580 align:middle line:84% say the population size, its growth being proportional. 00:02:41.580 --> 00:02:44.890 align:middle line:84% Then we add this extra term, 1 minus y, to-- 00:02:44.890 --> 00:02:47.790 align:middle line:84% to represent rather than just the simple exponential curve 00:02:47.790 --> 00:02:50.370 align:middle line:84% that you get at the beginning with a population 00:02:50.370 --> 00:02:52.200 align:middle line:90% growth in infinite resources. 00:02:52.200 --> 00:02:56.220 align:middle line:84% As you get close to the maximum resources here, one, 00:02:56.220 --> 00:02:58.680 align:middle line:84% you start to either plateau as you 00:02:58.680 --> 00:03:01.830 align:middle line:84% would if this were a ordinary differential equation, 00:03:01.830 --> 00:03:03.510 align:middle line:84% or we start to oscillate as you do 00:03:03.510 --> 00:03:06.630 align:middle line:84% this with an iterative solution, which we illustrated. 00:03:06.630 --> 00:03:10.240 align:middle line:84% And if the rate constant K gets large enough, 00:03:10.240 --> 00:03:15.130 align:middle line:84% then you get chaotic behavior where 00:03:15.130 --> 00:03:17.980 align:middle line:84% it can go close enough to zero that you're effectively 00:03:17.980 --> 00:03:22.487 align:middle line:84% simulating an extinction of the population. 00:03:22.487 --> 00:03:27.880 align:middle line:84% This-- the issues behind replication and approximation 00:03:27.880 --> 00:03:31.990 align:middle line:84% come together-- or came together in the concept of mutation, 00:03:31.990 --> 00:03:34.390 align:middle line:84% and in particular, the mutations that occur-- 00:03:34.390 --> 00:03:37.870 align:middle line:84% that might occur in single molecules, life being full 00:03:37.870 --> 00:03:40.600 align:middle line:84% of very important single molecules such as DNA. 00:03:40.600 --> 00:03:45.220 align:middle line:84% And here, even though you know there must be stochastics 00:03:45.220 --> 00:03:49.030 align:middle line:84% underlying this process because it is a single molecule, 00:03:49.030 --> 00:03:51.010 align:middle line:84% there has to be some way of overcoming it, 00:03:51.010 --> 00:03:53.177 align:middle line:84% because we went through a calculation that indicated 00:03:53.177 --> 00:03:57.580 align:middle line:84% that the 46 chromosomes in every one of your cells in your body 00:03:57.580 --> 00:03:59.650 align:middle line:84% has to be replicating quite faithfully, 00:03:59.650 --> 00:04:02.530 align:middle line:90% or else you'd be getting cancer. 00:04:02.530 --> 00:04:05.770 align:middle line:84% And that noise is overcome by the single moleculeness 00:04:05.770 --> 00:04:11.620 align:middle line:84% of the DNA being compensated by a multi-moleculeness of-- 00:04:11.620 --> 00:04:15.160 align:middle line:84% of energy containing molecules such as ATP 00:04:15.160 --> 00:04:19.269 align:middle line:84% and associated proteins and so forth. 00:04:19.269 --> 00:04:23.280 align:middle line:84% So now we've brought together the approximations 00:04:23.280 --> 00:04:27.955 align:middle line:90% and replication, and what this-- 00:04:27.955 --> 00:04:31.090 align:middle line:84% what replication leads to are pedigrees. 00:04:31.090 --> 00:04:34.300 align:middle line:84% Pedigrees are an example of many that we listed of 00:04:34.300 --> 00:04:37.690 align:middle line:84% directed acyclic graphs will have more pedigrees today. 00:04:37.690 --> 00:04:41.920 align:middle line:84% In a certain sense, the ultimate pedigree will be shown today. 00:04:41.920 --> 00:04:47.050 align:middle line:84% And the mutations also can be nicely modeled by-- 00:04:47.050 --> 00:04:48.940 align:middle line:84% depending on the exact application 00:04:48.940 --> 00:04:51.880 align:middle line:84% of your thought about mutations, by the binomial, the Poisson, 00:04:51.880 --> 00:04:55.900 align:middle line:84% and the normal distribution, the normal being continuous 00:04:55.900 --> 00:04:58.880 align:middle line:84% and the other two being discrete. 00:04:58.880 --> 00:05:01.910 align:middle line:84% And selection figured prominently-- and optimality 00:05:01.910 --> 00:05:03.380 align:middle line:90% figures prominently in biology. 00:05:03.380 --> 00:05:07.040 align:middle line:84% You typically won't lose a bet betting on optimality 00:05:07.040 --> 00:05:10.580 align:middle line:84% for at least exactly the circumstances 00:05:10.580 --> 00:05:14.090 align:middle line:84% under which the system has been subjected 00:05:14.090 --> 00:05:15.970 align:middle line:90% for millions of years. 00:05:15.970 --> 00:05:17.740 align:middle line:84% So this is the outline for today. 00:05:17.740 --> 00:05:20.730 align:middle line:84% Again, supposedly more on the biological side. 00:05:20.730 --> 00:05:25.410 align:middle line:84% We'll go through how purification 00:05:25.410 --> 00:05:29.340 align:middle line:84% has played a central role in the reductionist approach 00:05:29.340 --> 00:05:34.095 align:middle line:84% to biology and biochemistry, and how that purification is-- 00:05:34.095 --> 00:05:37.530 align:middle line:84% is also the antidote to the reductionism in that it 00:05:37.530 --> 00:05:39.840 align:middle line:84% provides a way of creating synthesis 00:05:39.840 --> 00:05:42.420 align:middle line:90% in going back up to systems. 00:05:42.420 --> 00:05:44.610 align:middle line:84% Systems biology is the second topic, 00:05:44.610 --> 00:05:49.440 align:middle line:84% and this is both relevant to the models, applications of models, 00:05:49.440 --> 00:05:53.250 align:middle line:84% and making the interconnections of the components 00:05:53.250 --> 00:05:58.650 align:middle line:84% ultimately in a synthetic loop of discovery and recreation 00:05:58.650 --> 00:06:00.780 align:middle line:90% perturbation. 00:06:00.780 --> 00:06:02.600 align:middle line:84% Then this is the ultimate pedigree 00:06:02.600 --> 00:06:03.680 align:middle line:90% that I was talking about. 00:06:03.680 --> 00:06:06.320 align:middle line:84% Continuity of life and how it applies 00:06:06.320 --> 00:06:08.270 align:middle line:84% to central dogma as the illustration of one 00:06:08.270 --> 00:06:10.100 align:middle line:84% of the most robust algorithms that we 00:06:10.100 --> 00:06:13.910 align:middle line:84% have in computational biology, which is the genetic code. 00:06:13.910 --> 00:06:17.630 align:middle line:84% A truly elegant discovery, and elegant 00:06:17.630 --> 00:06:19.400 align:middle line:90% partly because that's what-- 00:06:19.400 --> 00:06:23.760 align:middle line:84% the biology underlying it is so amazing. 00:06:23.760 --> 00:06:26.450 align:middle line:84% Then we go into the issues behind how 00:06:26.450 --> 00:06:30.020 align:middle line:84% we get qualitative models from quantitative data, 00:06:30.020 --> 00:06:32.780 align:middle line:84% and then how we go from those quantitative-- qualitative 00:06:32.780 --> 00:06:36.140 align:middle line:84% models and fill them again with the quantitation that's 00:06:36.140 --> 00:06:40.730 align:middle line:84% required for truth simulation, prediction, and design. 00:06:40.730 --> 00:06:43.430 align:middle line:84% And then we end up again on mutations of selection 00:06:43.430 --> 00:06:45.390 align:middle line:90% just as we did last time. 00:06:45.390 --> 00:06:46.340 align:middle line:90% OK. 00:06:46.340 --> 00:06:49.540 align:middle line:84% Familiar face-- the periodic table. 00:06:49.540 --> 00:06:51.860 align:middle line:84% Just like last time, the elements 00:06:51.860 --> 00:06:56.210 align:middle line:84% that are involved in the three major biopolymers in the cell. 00:06:56.210 --> 00:07:00.270 align:middle line:84% DNA goes to RNA, goes to protein is the central dogma, 00:07:00.270 --> 00:07:04.040 align:middle line:84% and all of those can be made of these six 00:07:04.040 --> 00:07:07.910 align:middle line:84% elements plus counterions of the polyanionic. 00:07:07.910 --> 00:07:11.420 align:middle line:84% So those are typically sodium, potassium, 00:07:11.420 --> 00:07:13.190 align:middle line:90% and other counterions. 00:07:13.190 --> 00:07:17.060 align:middle line:84% And so the total of 19 elements you can find in almost all life 00:07:17.060 --> 00:07:23.090 align:middle line:84% forms, and I would say many of the other elements on here 00:07:23.090 --> 00:07:26.660 align:middle line:84% that at least have some stable isotope in nature 00:07:26.660 --> 00:07:29.840 align:middle line:84% probably have some story that some living organism could tell 00:07:29.840 --> 00:07:32.480 align:middle line:90% if you were interested enough. 00:07:32.480 --> 00:07:35.090 align:middle line:84% And that might make a good thesis project 00:07:35.090 --> 00:07:37.610 align:middle line:84% or a good project for this class to talk about, 00:07:37.610 --> 00:07:41.060 align:middle line:84% what organisms do to either detoxify or to use 00:07:41.060 --> 00:07:46.790 align:middle line:84% in some exotic way each of the stable isotopes on this chart. 00:07:46.790 --> 00:07:52.460 align:middle line:84% Now, most organisms do not use these in the elemental form. 00:07:52.460 --> 00:07:54.590 align:middle line:84% They typically-- that in the middle here 00:07:54.590 --> 00:07:58.790 align:middle line:84% are elemental forms of these elements. 00:07:58.790 --> 00:08:02.060 align:middle line:84% Oxygen, hydrogen, and nitrogen are commonly 00:08:02.060 --> 00:08:08.010 align:middle line:84% used elemental forms, but there exists 00:08:08.010 --> 00:08:11.150 align:middle line:84% life forms for which none of these are really required. 00:08:11.150 --> 00:08:16.010 align:middle line:84% For example, oxygen is toxic to a variety of so called 00:08:16.010 --> 00:08:21.560 align:middle line:84% constitutive anaerobes, and nitrogen gas 00:08:21.560 --> 00:08:26.870 align:middle line:84% is only used by nitrogen fixing microorganisms. 00:08:26.870 --> 00:08:29.150 align:middle line:84% Each of these elements exist in a gaseous form. 00:08:29.150 --> 00:08:30.810 align:middle line:90% They tend to be more reduced-- 00:08:30.810 --> 00:08:33.980 align:middle line:84% that is to say, more hydrogen containing on the far left hand 00:08:33.980 --> 00:08:37.010 align:middle line:84% side of the slide, and they tend to be more oxidized 00:08:37.010 --> 00:08:40.280 align:middle line:84% on the right hand side, and in the oxidized form tend 00:08:40.280 --> 00:08:44.720 align:middle line:90% to be found often as salts. 00:08:44.720 --> 00:08:48.110 align:middle line:84% Here carbonate, for example, being the salt form 00:08:48.110 --> 00:08:50.960 align:middle line:84% of carbon dioxide in the oxidized form that's 00:08:50.960 --> 00:08:54.860 align:middle line:84% fixed by plants and photosynthetic bacteria 00:08:54.860 --> 00:09:00.200 align:middle line:84% in the oceans, and unfortunately in its CO2 form, 00:09:00.200 --> 00:09:02.570 align:middle line:90% major global warming gas. 00:09:02.570 --> 00:09:07.430 align:middle line:84% And you can see that every of the basic elements here has-- 00:09:07.430 --> 00:09:10.370 align:middle line:90% can be obtained as salts. 00:09:10.370 --> 00:09:14.510 align:middle line:84% Many organisms require much more complicated versions of this, 00:09:14.510 --> 00:09:19.481 align:middle line:84% require eating macromolecules like steak, and-- 00:09:19.481 --> 00:09:22.340 align:middle line:84% and maybe have even very much more exotic ones. 00:09:22.340 --> 00:09:29.510 align:middle line:84% As we purify-- when we represent the elements 00:09:29.510 --> 00:09:34.980 align:middle line:84% in that periodic table, those were the hard work 00:09:34.980 --> 00:09:38.970 align:middle line:84% of chemists that had to purify each of those elements. 00:09:38.970 --> 00:09:41.280 align:middle line:84% Not just as molecules, but it wasn't 00:09:41.280 --> 00:09:46.650 align:middle line:84% sufficient to get hydrocarbons, it had to get pure carbon. 00:09:46.650 --> 00:09:50.010 align:middle line:84% And then-- but that was only the starting point 00:09:50.010 --> 00:09:53.460 align:middle line:84% for determining not just the properties-- the elements-- 00:09:53.460 --> 00:09:55.690 align:middle line:84% but putting them back together as molecules. 00:09:55.690 --> 00:09:57.780 align:middle line:84% And this reduction down to elements and then 00:09:57.780 --> 00:10:01.470 align:middle line:84% re-synthesis back to molecules that was part of the proof 00:10:01.470 --> 00:10:04.060 align:middle line:84% that they really understood what the molecules were all about. 00:10:04.060 --> 00:10:07.380 align:middle line:84% And the same thing can be obtained in living systems. 00:10:07.380 --> 00:10:10.560 align:middle line:84% We typically start with molecules 00:10:10.560 --> 00:10:14.590 align:middle line:84% that are a covalent connection in some form or another. 00:10:14.590 --> 00:10:16.710 align:middle line:84% And when they non-covalently associate, 00:10:16.710 --> 00:10:18.030 align:middle line:90% they're called assemblies. 00:10:18.030 --> 00:10:21.060 align:middle line:84% Typically the assemblies that most biologists work with 00:10:21.060 --> 00:10:24.147 align:middle line:84% are assemblies of proteins or proteins and nucleic acids. 00:10:24.147 --> 00:10:25.980 align:middle line:84% But there are many, many kinds of assemblies 00:10:25.980 --> 00:10:29.520 align:middle line:84% and these assemblies blur in their-- 00:10:29.520 --> 00:10:33.270 align:middle line:84% in their definition into organelles. 00:10:33.270 --> 00:10:36.330 align:middle line:84% Organelles are basically assemblies 00:10:36.330 --> 00:10:38.280 align:middle line:84% that start getting large enough that they can 00:10:38.280 --> 00:10:40.890 align:middle line:90% be seen under simple micro-- 00:10:40.890 --> 00:10:44.730 align:middle line:90% microscopy scenarios. 00:10:44.730 --> 00:10:48.300 align:middle line:84% Organelles are not always, but are often bounded 00:10:48.300 --> 00:10:50.140 align:middle line:90% by lipid membranes. 00:10:50.140 --> 00:10:54.960 align:middle line:84% These are hydrocarbon containing bilayers or multilayers. 00:10:54.960 --> 00:10:57.540 align:middle line:84% Anyway, the distinction between macromolecular assemblies 00:10:57.540 --> 00:11:00.720 align:middle line:90% and organelles is soft. 00:11:00.720 --> 00:11:04.740 align:middle line:84% And then cells are collections of these assemblies, 00:11:04.740 --> 00:11:09.210 align:middle line:84% again typically bounded by a phospholipid bilayer. 00:11:09.210 --> 00:11:13.860 align:middle line:90% And some organisms are cells. 00:11:13.860 --> 00:11:18.420 align:middle line:84% Unicellular organisms are vast majority of organisms on Earth, 00:11:18.420 --> 00:11:19.770 align:middle line:90% and they are-- 00:11:19.770 --> 00:11:22.980 align:middle line:84% many of them are single cellular aggregates of cells. 00:11:22.980 --> 00:11:26.310 align:middle line:84% While multicellular organisms such as yours truly 00:11:26.310 --> 00:11:30.180 align:middle line:84% and everybody in this room can contain up 00:11:30.180 --> 00:11:35.340 align:middle line:84% to 10 to the 14th cells that are direct and regulated 00:11:35.340 --> 00:11:36.220 align:middle line:90% descendants. 00:11:36.220 --> 00:11:38.610 align:middle line:84% Now what are examples of purification methods 00:11:38.610 --> 00:11:41.460 align:middle line:90% that are actually used in-- 00:11:41.460 --> 00:11:43.830 align:middle line:84% on the road to computational biology? 00:11:43.830 --> 00:11:46.560 align:middle line:84% We have chromatography, electrophoresis, 00:11:46.560 --> 00:11:48.120 align:middle line:84% and sedimentation-- are very common 00:11:48.120 --> 00:11:51.510 align:middle line:84% ways of separating molecules, including protein molecules 00:11:51.510 --> 00:11:52.620 align:middle line:90% and assemblies. 00:11:52.620 --> 00:11:57.130 align:middle line:84% Actually organelles cells as well 00:11:57.130 --> 00:12:00.490 align:middle line:90% can be separated by these media. 00:12:00.490 --> 00:12:02.890 align:middle line:84% By far, one of the most common ways 00:12:02.890 --> 00:12:11.540 align:middle line:84% of separating these protein molecules 00:12:11.540 --> 00:12:14.480 align:middle line:84% is via chromatography and electrophoresis, 00:12:14.480 --> 00:12:19.100 align:middle line:84% and we'll see some examples in just a couple of slides. 00:12:19.100 --> 00:12:26.060 align:middle line:84% A very incredibly powerful way of separating entities 00:12:26.060 --> 00:12:29.600 align:middle line:84% in general is represented here on the far right hand side. 00:12:29.600 --> 00:12:31.670 align:middle line:90% This is clonal growth, or-- 00:12:31.670 --> 00:12:37.760 align:middle line:84% this is essentially an analysis of single molecules 00:12:37.760 --> 00:12:39.350 align:middle line:90% or single organisms. 00:12:39.350 --> 00:12:43.400 align:middle line:84% Each of these colonies, which might 00:12:43.400 --> 00:12:47.240 align:middle line:84% be growing on a Petri plate about the size of my fist, 00:12:47.240 --> 00:12:49.280 align:middle line:90% is a-- 00:12:49.280 --> 00:12:55.070 align:middle line:84% represents the growth from one starting cell to 10 00:12:55.070 --> 00:12:58.620 align:middle line:90% to the 8th or so final cells. 00:12:58.620 --> 00:13:00.830 align:middle line:84% Finally going through an exponential growth 00:13:00.830 --> 00:13:04.040 align:middle line:84% just as we went through in last class 00:13:04.040 --> 00:13:09.250 align:middle line:84% until the point that they have depleted 00:13:09.250 --> 00:13:13.630 align:middle line:84% the local resources near them, or have 00:13:13.630 --> 00:13:16.180 align:middle line:84% produced enough toxic waste products 00:13:16.180 --> 00:13:20.115 align:middle line:84% that they have slowed down their growth to form these colonies. 00:13:20.115 --> 00:13:21.490 align:middle line:84% In certain organisms they'll just 00:13:21.490 --> 00:13:24.760 align:middle line:84% keep growing until they get to the edges of the Petri dish, 00:13:24.760 --> 00:13:27.700 align:middle line:84% but this is much more general than just growing 00:13:27.700 --> 00:13:29.920 align:middle line:90% bacterial colonies. 00:13:29.920 --> 00:13:35.410 align:middle line:84% You can see it, almost any organism that 00:13:35.410 --> 00:13:39.490 align:middle line:84% has a limited ability-- limited motility, 00:13:39.490 --> 00:13:43.000 align:middle line:84% these will form little clones such as-- this happens 00:13:43.000 --> 00:13:45.910 align:middle line:84% in various tree populations, for example. 00:13:45.910 --> 00:13:49.330 align:middle line:84% Also it represents the ultimate purification. 00:13:49.330 --> 00:13:52.855 align:middle line:84% In one step, you're getting something 00:13:52.855 --> 00:13:56.710 align:middle line:84% that would by combinations of different chromatographic steps 00:13:56.710 --> 00:13:59.650 align:middle line:84% and electrophoretic and sedimentation steps-- 00:13:59.650 --> 00:14:03.730 align:middle line:84% you might have to serially do several of these steps in a row 00:14:03.730 --> 00:14:07.630 align:middle line:84% to get a molecule or assembly purified. 00:14:07.630 --> 00:14:11.810 align:middle line:84% While here in one step by essentially limiting dilution-- 00:14:11.810 --> 00:14:14.100 align:middle line:90% you dilute the-- 00:14:14.100 --> 00:14:16.480 align:middle line:84% the molecule of interest to the point 00:14:16.480 --> 00:14:18.140 align:middle line:90% where it's a single molecule. 00:14:18.140 --> 00:14:20.037 align:middle line:84% Well now if they don't undergo clonal growth, 00:14:20.037 --> 00:14:22.120 align:middle line:84% this is not terribly useful because it's very hard 00:14:22.120 --> 00:14:24.910 align:middle line:84% to study single molecules, even if they're well 00:14:24.910 --> 00:14:27.820 align:middle line:84% isolated from all the contaminating molecules. 00:14:27.820 --> 00:14:30.370 align:middle line:84% There are ways, and we will talk about them in the course, 00:14:30.370 --> 00:14:35.728 align:middle line:84% but you need-- ideally you need some way of amplifying them. 00:14:35.728 --> 00:14:40.300 align:middle line:84% There are ways to amplify nucleic acid molecules such 00:14:40.300 --> 00:14:43.785 align:middle line:84% that they exhibit clonal growth like this. 00:14:43.785 --> 00:14:46.090 align:middle line:90% And in principle, any-- 00:14:46.090 --> 00:14:49.360 align:middle line:84% either by putting them into a bacterium 00:14:49.360 --> 00:14:52.990 align:middle line:84% so the bacterium behaves in this manner, carrying along with it 00:14:52.990 --> 00:14:55.780 align:middle line:84% the artificial piece of nucleic acid you're interested. 00:14:55.780 --> 00:14:58.690 align:middle line:84% Or you can do it entirely with enzymes 00:14:58.690 --> 00:15:01.120 align:middle line:84% so that the nucleic acids replicate 00:15:01.120 --> 00:15:05.570 align:middle line:84% and make these colony like objects. 00:15:05.570 --> 00:15:11.410 align:middle line:84% Now this is an idiosyncratic view of this purification-- 00:15:11.410 --> 00:15:15.190 align:middle line:84% of this process by which we as scientists have gone through 00:15:15.190 --> 00:15:18.970 align:middle line:84% purification and then are returning to much less pure 00:15:18.970 --> 00:15:22.276 align:middle line:90% systems as a subject of our-- 00:15:22.276 --> 00:15:24.160 align:middle line:90% of our research. 00:15:24.160 --> 00:15:27.880 align:middle line:84% This is-- we'll call this three revolutions. 00:15:27.880 --> 00:15:33.670 align:middle line:84% In the pre-1970s, we had column chromatography-- so 00:15:33.670 --> 00:15:35.380 align:middle line:84% called chromatography in that last slide, 00:15:35.380 --> 00:15:37.810 align:middle line:84% because literally the substances that were being separated 00:15:37.810 --> 00:15:42.400 align:middle line:84% were highly colored as were those bands in the last slide, 00:15:42.400 --> 00:15:45.000 align:middle line:90% those two dark bands. 00:15:45.000 --> 00:15:46.510 align:middle line:90% Hence the name chromatography. 00:15:46.510 --> 00:15:52.080 align:middle line:84% It's really a separation by the properties of the solid phase 00:15:52.080 --> 00:15:54.360 align:middle line:84% and the properties of the mobile phase, 00:15:54.360 --> 00:15:56.790 align:middle line:84% and the molecules in the mobile phase being 00:15:56.790 --> 00:15:59.850 align:middle line:84% separated by differentiable absorption to the solid phase. 00:15:59.850 --> 00:16:04.185 align:middle line:84% Gel electrophoresis and sedimentation 00:16:04.185 --> 00:16:06.480 align:middle line:90% in a gravitational field-- 00:16:06.480 --> 00:16:09.480 align:middle line:84% these were all part of this amazing revolution 00:16:09.480 --> 00:16:10.740 align:middle line:90% that allowed us-- 00:16:10.740 --> 00:16:15.240 align:middle line:84% allowed scientists to get molecules, assemblies, 00:16:15.240 --> 00:16:21.730 align:middle line:84% and cells purified away from other contaminants. 00:16:21.730 --> 00:16:23.848 align:middle line:84% Then recombinant DNA did that trick 00:16:23.848 --> 00:16:26.390 align:middle line:84% that was in the lower right hand slide of the previous slide, 00:16:26.390 --> 00:16:30.770 align:middle line:84% which was going directly to purification 00:16:30.770 --> 00:16:35.300 align:middle line:84% by going to single molecule isolation 00:16:35.300 --> 00:16:36.920 align:middle line:84% by dilution to the point where you had 00:16:36.920 --> 00:16:40.190 align:middle line:84% less than one molecule per cell, and less 00:16:40.190 --> 00:16:43.790 align:middle line:84% than one cell per square centimeter on the Petri plate. 00:16:43.790 --> 00:16:45.470 align:middle line:84% That gives you single step purity 00:16:45.470 --> 00:16:48.290 align:middle line:84% of the gene, which in effect allows 00:16:48.290 --> 00:16:51.050 align:middle line:84% you to get single step purity of whatever 00:16:51.050 --> 00:16:52.460 align:middle line:90% is encoded by the gene-- 00:16:52.460 --> 00:16:56.330 align:middle line:84% the RNA, the protein, or the enzymatic activity. 00:16:56.330 --> 00:16:58.480 align:middle line:90% This was a huge change. 00:16:58.480 --> 00:17:03.040 align:middle line:84% Suddenly everybody was spending a lot of time so called 00:17:03.040 --> 00:17:07.700 align:middle line:84% cloning of DNA and sequencing it, 00:17:07.700 --> 00:17:11.141 align:middle line:84% and almost every thesis and paper of that time-- 00:17:11.141 --> 00:17:13.099 align:middle line:84% everybody was turning into molecular biologists 00:17:13.099 --> 00:17:16.730 align:middle line:84% in order to do this, and it became very routine and very 00:17:16.730 --> 00:17:18.470 align:middle line:90% time consuming and expensive. 00:17:18.470 --> 00:17:22.280 align:middle line:84% And so the third revolution was automating this 00:17:22.280 --> 00:17:27.200 align:middle line:84% and using economies of scale so that all the genes were 00:17:27.200 --> 00:17:30.860 align:middle line:84% obtained at once and sequenced at once, 00:17:30.860 --> 00:17:32.423 align:middle line:84% rather than going through here where 00:17:32.423 --> 00:17:35.090 align:middle line:84% you would have to go through the entire library of all the genes 00:17:35.090 --> 00:17:37.280 align:middle line:84% just to find your favorite one, and then you 00:17:37.280 --> 00:17:41.120 align:middle line:84% would sequence that one working hard on isolating it away 00:17:41.120 --> 00:17:42.120 align:middle line:90% from everything else. 00:17:42.120 --> 00:17:44.120 align:middle line:84% But with sequencing genomes, it was more 00:17:44.120 --> 00:17:48.210 align:middle line:84% a process of everything you came upon was interesting, 00:17:48.210 --> 00:17:50.960 align:middle line:84% and so you didn't have to do quite as much time selecting, 00:17:50.960 --> 00:17:54.840 align:middle line:84% you just made it more of a production effort. 00:17:54.840 --> 00:18:00.450 align:middle line:84% But the subtext of this was not just automation and-- 00:18:00.450 --> 00:18:02.190 align:middle line:90% and economies of scale. 00:18:02.190 --> 00:18:05.370 align:middle line:84% It also started to return us to thinking about whole systems 00:18:05.370 --> 00:18:06.990 align:middle line:90% and doing things systematically. 00:18:06.990 --> 00:18:12.990 align:middle line:84% And this was particularly valuable in the sequelae 00:18:12.990 --> 00:18:15.270 align:middle line:84% of genome sequencing, which was functional 00:18:15.270 --> 00:18:19.000 align:middle line:84% genomics, which we'll talk about quite a bit in this course-- 00:18:19.000 --> 00:18:24.120 align:middle line:84% applying the same attitude to other biological measurements, 00:18:24.120 --> 00:18:25.930 align:middle line:84% and this returns us to whole systems. 00:18:25.930 --> 00:18:29.850 align:middle line:84% Now that leads us to the discussion of whole systems 00:18:29.850 --> 00:18:30.780 align:middle line:90% modeling-- 00:18:30.780 --> 00:18:35.240 align:middle line:90% 00:18:35.240 --> 00:18:37.760 align:middle line:84% systems biology and the models therein. 00:18:37.760 --> 00:18:43.850 align:middle line:84% So we have-- this is just one of the earlier papers. 00:18:43.850 --> 00:18:46.646 align:middle line:84% There are many now that are trying to-- 00:18:46.646 --> 00:18:48.530 align:middle line:84% we're trying to grope our way towards what 00:18:48.530 --> 00:18:51.660 align:middle line:84% we mean by system biology, but this is 00:18:51.660 --> 00:18:54.080 align:middle line:90% paraphrasing from that paper. 00:18:54.080 --> 00:18:56.600 align:middle line:84% We want to follow these four steps as a protocol 00:18:56.600 --> 00:18:59.060 align:middle line:84% to find all the components of the system. 00:18:59.060 --> 00:19:01.280 align:middle line:84% Systematically perturb and monitor 00:19:01.280 --> 00:19:04.880 align:middle line:84% the components of the system so that we 00:19:04.880 --> 00:19:07.730 align:middle line:84% can do this either genetically or environmentally, meaning 00:19:07.730 --> 00:19:12.080 align:middle line:84% changing the small and large molecules that program the-- 00:19:12.080 --> 00:19:14.720 align:middle line:84% the biological system from the outside. 00:19:14.720 --> 00:19:18.590 align:middle line:84% Then refine the model which you had 00:19:18.590 --> 00:19:23.690 align:middle line:84% maybe before perturbing it such that the predictions most 00:19:23.690 --> 00:19:27.900 align:middle line:90% closely agree with observations. 00:19:27.900 --> 00:19:30.330 align:middle line:84% Listen carefully to that statement-- 00:19:30.330 --> 00:19:31.920 align:middle line:84% refine the model so that predictions 00:19:31.920 --> 00:19:33.480 align:middle line:84% agree with observations, and then 00:19:33.480 --> 00:19:37.410 align:middle line:84% do new perturbation experiments to distinguish among the model 00:19:37.410 --> 00:19:38.670 align:middle line:90% hypotheses. 00:19:38.670 --> 00:19:43.500 align:middle line:84% We do this in a cyclic fashion, basically 00:19:43.500 --> 00:19:45.660 align:middle line:84% going back up to item two so that we're 00:19:45.660 --> 00:19:47.550 align:middle line:90% perturbing and monitoring. 00:19:47.550 --> 00:19:51.760 align:middle line:84% Now what is the-- what's the critique of this systems 00:19:51.760 --> 00:19:54.250 align:middle line:90% biology manifesto? 00:19:54.250 --> 00:19:58.760 align:middle line:84% We have-- those of you who have read 00:19:58.760 --> 00:20:01.460 align:middle line:84% books that predate the genome project and systems 00:20:01.460 --> 00:20:05.705 align:middle line:84% biology say, hey, this-- what's new here? 00:20:05.705 --> 00:20:07.580 align:middle line:84% This is the way biologists were doing it even 00:20:07.580 --> 00:20:09.450 align:middle line:90% before recombinant DNA. 00:20:09.450 --> 00:20:13.490 align:middle line:84% So it is an old approach, but the new spin on it is that-- 00:20:13.490 --> 00:20:15.230 align:middle line:90% is the word "all components." 00:20:15.230 --> 00:20:18.380 align:middle line:84% Typically before, the components would be chosen 00:20:18.380 --> 00:20:21.080 align:middle line:84% and the perturbations would be chosen 00:20:21.080 --> 00:20:23.540 align:middle line:84% based on the latest biological fad 00:20:23.540 --> 00:20:25.640 align:middle line:84% or what was available technically 00:20:25.640 --> 00:20:29.540 align:middle line:84% at the time based on the history of the component studies 00:20:29.540 --> 00:20:30.230 align:middle line:90% before that. 00:20:30.230 --> 00:20:34.790 align:middle line:84% So it's a significant deviation to now even set 00:20:34.790 --> 00:20:36.140 align:middle line:90% as a goal all components. 00:20:36.140 --> 00:20:39.050 align:middle line:84% A very challenging goal is been met 00:20:39.050 --> 00:20:41.510 align:middle line:84% in the case of certain genome sequences, 00:20:41.510 --> 00:20:44.840 align:middle line:84% it has not been met in any functional genomics 00:20:44.840 --> 00:20:45.818 align:middle line:90% that I'm aware of. 00:20:45.818 --> 00:20:47.360 align:middle line:84% But it certainly is a goal, and we're 00:20:47.360 --> 00:20:49.100 align:middle line:84% getting asymptotically close to it 00:20:49.100 --> 00:20:52.160 align:middle line:84% just as we got asymptotically close to the genome sequences. 00:20:52.160 --> 00:20:55.250 align:middle line:84% To systematically perturb has the conceit 00:20:55.250 --> 00:20:59.605 align:middle line:84% that we can list all the perturbations 00:20:59.605 --> 00:21:00.980 align:middle line:84% we would want to do and then walk 00:21:00.980 --> 00:21:05.050 align:middle line:84% through them in a systematic way rather than a-- 00:21:05.050 --> 00:21:08.200 align:middle line:90% a more whimsical way. 00:21:08.200 --> 00:21:15.270 align:middle line:84% So this is new spins, but what's missing from this manifesto 00:21:15.270 --> 00:21:17.310 align:middle line:90% in the previous slide that-- 00:21:17.310 --> 00:21:18.690 align:middle line:90% in systems biology? 00:21:18.690 --> 00:21:23.040 align:middle line:84% For one thing, and I cautioned you in the previous slide 00:21:23.040 --> 00:21:25.770 align:middle line:84% that when you start fitting your model to the data 00:21:25.770 --> 00:21:28.800 align:middle line:84% and refining your model, there's a problem of overfitting. 00:21:28.800 --> 00:21:31.600 align:middle line:84% And this will come up a couple of times in-- 00:21:31.600 --> 00:21:33.880 align:middle line:90% in this course. 00:21:33.880 --> 00:21:35.880 align:middle line:84% If you have enough adjustable parameters, 00:21:35.880 --> 00:21:37.710 align:middle line:84% you can fit almost anything, and so you 00:21:37.710 --> 00:21:41.160 align:middle line:84% have to be careful that as you refine your model that you 00:21:41.160 --> 00:21:43.830 align:middle line:84% state exactly how many adjustable parameters you have 00:21:43.830 --> 00:21:44.770 align:middle line:90% and-- 00:21:44.770 --> 00:21:48.030 align:middle line:84% and how many data points you have that 00:21:48.030 --> 00:21:49.830 align:middle line:84% are truly independent for fitting that. 00:21:49.830 --> 00:21:53.460 align:middle line:84% So we have methods to recapture unautomated data. 00:21:53.460 --> 00:21:57.290 align:middle line:90% 00:21:57.290 --> 00:22:00.260 align:middle line:84% There is a step implicit in the previous one-- 00:22:00.260 --> 00:22:02.745 align:middle line:84% actually explicitly stated elsewhere in some of these 00:22:02.745 --> 00:22:03.245 align:middle line:90% papers-- 00:22:03.245 --> 00:22:05.825 align:middle line:90% 00:22:05.825 --> 00:22:09.536 align:middle line:90% that when you have-- 00:22:09.536 --> 00:22:11.600 align:middle line:84% as you're developing the model, you 00:22:11.600 --> 00:22:15.590 align:middle line:84% will draw not only upon the systematically collected data, 00:22:15.590 --> 00:22:17.720 align:middle line:90% but also upon the literature. 00:22:17.720 --> 00:22:21.530 align:middle line:84% And the literature, as we'll see in the next couple of slides, 00:22:21.530 --> 00:22:24.770 align:middle line:84% not only has unautomated data, but it has models that 00:22:24.770 --> 00:22:26.583 align:middle line:90% are derived in a variety of-- 00:22:26.583 --> 00:22:29.361 align:middle line:90% 00:22:29.361 --> 00:22:33.790 align:middle line:84% of somewhat undisciplined-- or a different discipline. 00:22:33.790 --> 00:22:37.170 align:middle line:84% It's not an electronically compatible discipline. 00:22:37.170 --> 00:22:39.260 align:middle line:84% And so there's a process by which 00:22:39.260 --> 00:22:41.870 align:middle line:84% one captures this unautomated data 00:22:41.870 --> 00:22:43.520 align:middle line:84% and integrates it with the automated 00:22:43.520 --> 00:22:51.940 align:middle line:84% data, which can be either challenging or pathological. 00:22:51.940 --> 00:22:56.350 align:middle line:84% So we need to make more explicit these-- the logical connections 00:22:56.350 --> 00:23:01.480 align:middle line:84% that are used for deriving these systems biology diagrams 00:23:01.480 --> 00:23:05.020 align:middle line:90% and quantitative models. 00:23:05.020 --> 00:23:07.540 align:middle line:84% Finally, when you make these perturbation experiments, 00:23:07.540 --> 00:23:10.315 align:middle line:90% if they're done using-- 00:23:10.315 --> 00:23:12.220 align:middle line:84% there's a new optimization that needs 00:23:12.220 --> 00:23:15.790 align:middle line:84% to be made in order to integrate them with the systems biology 00:23:15.790 --> 00:23:18.010 align:middle line:90% loop. 00:23:18.010 --> 00:23:20.890 align:middle line:84% As I mentioned in the previous talk, 00:23:20.890 --> 00:23:23.200 align:middle line:84% the thing that makes the killer applications 00:23:23.200 --> 00:23:28.300 align:middle line:84% in computational biology so far are searching, merging, 00:23:28.300 --> 00:23:28.900 align:middle line:90% and checking. 00:23:28.900 --> 00:23:33.760 align:middle line:84% If you can find ways to-- to search, merge, and check 00:23:33.760 --> 00:23:37.450 align:middle line:84% large data sets via models, then you've 00:23:37.450 --> 00:23:38.980 align:middle line:90% made a great deal of progress. 00:23:38.980 --> 00:23:42.260 align:middle line:84% And that should be the goal here too. 00:23:42.260 --> 00:23:45.290 align:middle line:84% So what the systems biology will do, 00:23:45.290 --> 00:23:47.510 align:middle line:84% I think possibly more well illustrated 00:23:47.510 --> 00:23:52.200 align:middle line:84% by this slide and the last talk than by some of the examples 00:23:52.200 --> 00:23:54.410 align:middle line:84% so far in the literature on system biology-- 00:23:54.410 --> 00:23:56.660 align:middle line:84% but the goal is to be able to work 00:23:56.660 --> 00:24:00.080 align:middle line:84% with very simple parts, this reduction down 00:24:00.080 --> 00:24:03.830 align:middle line:84% to the basic parts, and then move up through models that 00:24:03.830 --> 00:24:08.210 align:middle line:84% are hierarchical and include all these intermediate steps 00:24:08.210 --> 00:24:12.830 align:middle line:84% to very high level ways of describing, understanding 00:24:12.830 --> 00:24:13.400 align:middle line:90% a system. 00:24:13.400 --> 00:24:16.160 align:middle line:84% This one you have the unfair advantage that the entire thing 00:24:16.160 --> 00:24:18.590 align:middle line:84% was designed from scratch by humans, 00:24:18.590 --> 00:24:20.360 align:middle line:84% but in biological systems you want-- 00:24:20.360 --> 00:24:22.370 align:middle line:84% you want to reverse engineer it to the point 00:24:22.370 --> 00:24:25.100 align:middle line:84% where it has some of the same flavor of [? board ?] 00:24:25.100 --> 00:24:29.450 align:middle line:84% engineered systems, and then develop 00:24:29.450 --> 00:24:32.450 align:middle line:84% ways of simulating the systems so that you 00:24:32.450 --> 00:24:35.180 align:middle line:90% can design new versions of it. 00:24:35.180 --> 00:24:38.630 align:middle line:84% Whenever you find yourself doing an experiment or computational 00:24:38.630 --> 00:24:41.160 align:middle line:84% biology, you're going to be asking yourself, 00:24:41.160 --> 00:24:42.980 align:middle line:90% why am I doing this? 00:24:42.980 --> 00:24:44.390 align:middle line:90% Why am I in this classroom? 00:24:44.390 --> 00:24:49.220 align:middle line:84% And whenever you do that, you should ask yourself 00:24:49.220 --> 00:24:53.210 align:middle line:84% why a bunch of times until you get down to the real core 00:24:53.210 --> 00:24:54.330 align:middle line:90% reasons. 00:24:54.330 --> 00:24:59.090 align:middle line:84% And so for example, we were in effect sequencing the genome 00:24:59.090 --> 00:25:01.040 align:middle line:90% prior to the genome project. 00:25:01.040 --> 00:25:04.010 align:middle line:84% We were spending hundreds of millions of dollars 00:25:04.010 --> 00:25:05.660 align:middle line:90% of NIH money and other-- 00:25:05.660 --> 00:25:09.810 align:middle line:84% every funding organization's money doing it 00:25:09.810 --> 00:25:11.490 align:middle line:90% in a very inefficient way. 00:25:11.490 --> 00:25:13.560 align:middle line:84% And we knew why we were doing it, I think. 00:25:13.560 --> 00:25:16.620 align:middle line:84% We wanted to map variation in sequences, 00:25:16.620 --> 00:25:18.343 align:middle line:84% variation within a species like humans 00:25:18.343 --> 00:25:20.010 align:middle line:84% that make us different from one another, 00:25:20.010 --> 00:25:22.480 align:middle line:84% variation between different species, 00:25:22.480 --> 00:25:25.560 align:middle line:84% which is comparative genomics, item three. 00:25:25.560 --> 00:25:27.420 align:middle line:84% And in between item two, we wanted 00:25:27.420 --> 00:25:31.950 align:middle line:84% to have a complete set of human RNAs, proteins, 00:25:31.950 --> 00:25:34.500 align:middle line:84% and regulatory elements, and for every other organism too. 00:25:34.500 --> 00:25:36.090 align:middle line:84% I just use human as an illustration 00:25:36.090 --> 00:25:39.060 align:middle line:84% because it was called the Human Genome Project. 00:25:39.060 --> 00:25:41.490 align:middle line:84% And we wanted this complete set so 00:25:41.490 --> 00:25:45.180 align:middle line:84% that we could go back and measure them systematically. 00:25:45.180 --> 00:25:47.212 align:middle line:84% Although this was not articulated 00:25:47.212 --> 00:25:50.100 align:middle line:84% in any-- in any way, it was-- we never 00:25:50.100 --> 00:25:53.320 align:middle line:84% did-- we didn't use the words complete or systematic 00:25:53.320 --> 00:25:56.650 align:middle line:84% very much prior to the Genome Project. 00:25:56.650 --> 00:25:58.510 align:middle line:84% And if we said these were the reasons, then 00:25:58.510 --> 00:26:01.000 align:middle line:84% we could ask why do we want to map variation, 00:26:01.000 --> 00:26:04.590 align:middle line:84% why do we want a partial or complete set 00:26:04.590 --> 00:26:08.910 align:middle line:84% of these various molecules and regulatory elements? 00:26:08.910 --> 00:26:12.550 align:middle line:84% And why do we want to compare them in different organisms 00:26:12.550 --> 00:26:15.540 align:middle line:90% and in different environments? 00:26:15.540 --> 00:26:17.460 align:middle line:84% And the answer to that would be that we 00:26:17.460 --> 00:26:21.102 align:middle line:84% would want to make quantitative biosystems models, 00:26:21.102 --> 00:26:22.560 align:middle line:84% such as the ones we were describing 00:26:22.560 --> 00:26:24.900 align:middle line:84% the last couple of slides, of the molecular interactions 00:26:24.900 --> 00:26:28.380 align:middle line:84% at all the levels extending from atoms to cells 00:26:28.380 --> 00:26:30.750 align:middle line:84% to organisms and populations of organisms, 00:26:30.750 --> 00:26:33.990 align:middle line:84% because it's the population upon which selection acts, 00:26:33.990 --> 00:26:37.770 align:middle line:84% and it's the population that allows us to understand 00:26:37.770 --> 00:26:39.970 align:middle line:90% and make useful products. 00:26:39.970 --> 00:26:41.460 align:middle line:84% And so when we ask, why do we want 00:26:41.460 --> 00:26:44.250 align:middle line:84% to make these biosystems models we have three reasons, some 00:26:44.250 --> 00:26:45.210 align:middle line:90% of which we touched-- 00:26:45.210 --> 00:26:47.970 align:middle line:84% all of which we touched upon last time when we asked, 00:26:47.970 --> 00:26:48.970 align:middle line:90% why do we model? 00:26:48.970 --> 00:26:51.540 align:middle line:84% Same thing, why are we collecting all this data 00:26:51.540 --> 00:26:52.260 align:middle line:90% to model? 00:26:52.260 --> 00:26:54.970 align:middle line:84% We model so that we can share information 00:26:54.970 --> 00:27:00.670 align:middle line:84% and so that we can construct a test of understanding. 00:27:00.670 --> 00:27:02.170 align:middle line:84% One of the tests of understanding 00:27:02.170 --> 00:27:03.520 align:middle line:90% is making useful products. 00:27:03.520 --> 00:27:06.320 align:middle line:90% 00:27:06.320 --> 00:27:12.200 align:middle line:84% So on that theme of why and making useful products, 00:27:12.200 --> 00:27:15.800 align:middle line:84% I will put this in the context of, say, 00:27:15.800 --> 00:27:18.200 align:middle line:84% the projects that you'll be doing for this class. 00:27:18.200 --> 00:27:21.470 align:middle line:84% I like to stimulate you to think about this early on, 00:27:21.470 --> 00:27:24.080 align:middle line:84% In every class I'll mention something. 00:27:24.080 --> 00:27:26.480 align:middle line:84% And you might say, well grand challenges-- 00:27:26.480 --> 00:27:30.890 align:middle line:84% grand and useful challenges are really not a great place 00:27:30.890 --> 00:27:35.090 align:middle line:84% to be doing a short term project for a course. 00:27:35.090 --> 00:27:37.820 align:middle line:84% But actually I think that the piece that you choose 00:27:37.820 --> 00:27:39.740 align:middle line:84% should be a piece of a grand challenge 00:27:39.740 --> 00:27:41.820 align:middle line:84% because it really gives you the context. 00:27:41.820 --> 00:27:44.300 align:middle line:84% And so I'm just going to walk through these three 00:27:44.300 --> 00:27:49.640 align:middle line:84% classes of challenge, not so you will feel limited but so 00:27:49.640 --> 00:27:54.200 align:middle line:84% that you will feel broadened in your mandate for what 00:27:54.200 --> 00:27:55.340 align:middle line:90% you can do. 00:27:55.340 --> 00:27:59.450 align:middle line:84% So at the simplest level, kind of reflecting last lecture, 00:27:59.450 --> 00:28:03.830 align:middle line:84% simple going from atoms to small cells 00:28:03.830 --> 00:28:08.630 align:middle line:84% with small genomes, maybe even minimal or miniature. 00:28:08.630 --> 00:28:13.280 align:middle line:84% Just like you want to downscale electronic components, 00:28:13.280 --> 00:28:17.800 align:middle line:84% we have, if we really want to show that we understand 00:28:17.800 --> 00:28:20.870 align:middle line:84% the biosynthetic route that will be the topic of today, 00:28:20.870 --> 00:28:24.950 align:middle line:84% this really key mechanism by which we go from DNA 00:28:24.950 --> 00:28:28.050 align:middle line:84% to proteins, we should be able to take it apart, 00:28:28.050 --> 00:28:29.210 align:middle line:90% put it back together again. 00:28:29.210 --> 00:28:31.627 align:middle line:84% That's the way-- that's one way of proving that we really 00:28:31.627 --> 00:28:32.210 align:middle line:90% understand it. 00:28:32.210 --> 00:28:36.620 align:middle line:84% That has not been done in a purely synthetic route. 00:28:36.620 --> 00:28:39.590 align:middle line:84% We have taken apart protein synthetic apparatus 00:28:39.590 --> 00:28:42.020 align:middle line:84% and put it back together again, but we have not completely 00:28:42.020 --> 00:28:44.940 align:middle line:84% synthesized the synthetic apparatus. 00:28:44.940 --> 00:28:50.090 align:middle line:84% That sounds odd, but that would be one step, 00:28:50.090 --> 00:28:52.640 align:middle line:84% but the impact of that would not be just proving 00:28:52.640 --> 00:28:53.630 align:middle line:90% that we can do it. 00:28:53.630 --> 00:28:56.480 align:middle line:84% The impact would be that we can now 00:28:56.480 --> 00:29:00.530 align:middle line:84% make a simple biological system that is self-replicating, uses 00:29:00.530 --> 00:29:05.420 align:middle line:84% proteins, and allows us to link the atomic changes 00:29:05.420 --> 00:29:06.950 align:middle line:90% to population evolution. 00:29:06.950 --> 00:29:10.850 align:middle line:84% In populations of humans, this would be daunting 00:29:10.850 --> 00:29:11.700 align:middle line:90% computationally. 00:29:11.700 --> 00:29:13.075 align:middle line:84% But when you think of populations 00:29:13.075 --> 00:29:15.260 align:middle line:84% of self-replicating molecules such as the simplest 00:29:15.260 --> 00:29:19.460 align:middle line:84% one last week, which is these trinucleotides being ligated 00:29:19.460 --> 00:29:21.470 align:middle line:84% into hexanucleotides, you can start 00:29:21.470 --> 00:29:25.073 align:middle line:84% to conceive of actually connecting the atomic modeling 00:29:25.073 --> 00:29:26.990 align:middle line:84% to the population modeling, which is basically 00:29:26.990 --> 00:29:28.260 align:middle line:90% the breadth of this course. 00:29:28.260 --> 00:29:30.500 align:middle line:84% It covers the whole breadth, it was collapsed down 00:29:30.500 --> 00:29:31.910 align:middle line:90% to that simple model. 00:29:31.910 --> 00:29:34.710 align:middle line:84% But more-- even more importantly, 00:29:34.710 --> 00:29:36.980 align:middle line:84% we can start engineering smart materials. 00:29:36.980 --> 00:29:39.830 align:middle line:84% Materials that have important properties 00:29:39.830 --> 00:29:44.630 align:middle line:84% that in a certain sense compute in chemistry. 00:29:44.630 --> 00:29:49.360 align:middle line:84% We can make a whole alternative chemistry 00:29:49.360 --> 00:29:53.220 align:middle line:84% of stereospecific meaning sensitive 00:29:53.220 --> 00:29:56.070 align:middle line:84% to the actual handedness of the molecules, which 00:29:56.070 --> 00:29:58.710 align:middle line:84% is so critical in pharmaceuticals 00:29:58.710 --> 00:30:00.760 align:middle line:90% and in enzymatics in general. 00:30:00.760 --> 00:30:02.520 align:middle line:84% This can be engineered by getting 00:30:02.520 --> 00:30:06.830 align:middle line:84% a handle on the synthetic machinery of life. 00:30:06.830 --> 00:30:10.030 align:middle line:84% OK, so that's one that's at the simple end of the spectrum. 00:30:10.030 --> 00:30:13.065 align:middle line:90% What about going from-- 00:30:13.065 --> 00:30:14.440 align:middle line:84% that's going from atoms to cells, 00:30:14.440 --> 00:30:16.430 align:middle line:90% how about cells to tissues? 00:30:16.430 --> 00:30:20.480 align:middle line:84% When we typically-- and many of you 00:30:20.480 --> 00:30:24.440 align:middle line:84% are either already in the biotech pharmaceutical industry 00:30:24.440 --> 00:30:26.300 align:middle line:84% or feel that the research that you might 00:30:26.300 --> 00:30:29.300 align:middle line:84% be doing as a graduate student would contribute to that 00:30:29.300 --> 00:30:32.030 align:middle line:90% in some way, however indirect. 00:30:32.030 --> 00:30:37.010 align:middle line:84% The way that you would program a computer, 00:30:37.010 --> 00:30:39.230 align:middle line:84% we might fill up this room full of laptops 00:30:39.230 --> 00:30:42.080 align:middle line:84% and go pouring random chemicals on them 00:30:42.080 --> 00:30:44.450 align:middle line:84% to see if they then produce the graphical user 00:30:44.450 --> 00:30:46.040 align:middle line:90% interface that we desire. 00:30:46.040 --> 00:30:52.350 align:middle line:84% This would be the drug screening approach to programming. 00:30:52.350 --> 00:30:54.560 align:middle line:84% And obviously the way we program actually 00:30:54.560 --> 00:30:59.720 align:middle line:84% is we work in the natural biopolymers of computers, which 00:30:59.720 --> 00:31:04.610 align:middle line:84% is the strings of zeros and ones represented in the computer. 00:31:04.610 --> 00:31:07.640 align:middle line:84% And we program those as long strings, 00:31:07.640 --> 00:31:10.040 align:middle line:84% and so the equivalent in cells if you-- 00:31:10.040 --> 00:31:13.550 align:middle line:84% might be to manipulate the genome itself. 00:31:13.550 --> 00:31:15.920 align:middle line:84% We're doing genome level programming either 00:31:15.920 --> 00:31:21.600 align:middle line:84% at DNA, RNA level via nucleic acids. 00:31:21.600 --> 00:31:23.360 align:middle line:84% This is not an either/or, this is probably 00:31:23.360 --> 00:31:25.280 align:middle line:84% something where one is augmenting 00:31:25.280 --> 00:31:29.240 align:middle line:84% by studying and programming at this detail level. 00:31:29.240 --> 00:31:33.170 align:middle line:84% And to do this manipulation of stem cells 00:31:33.170 --> 00:31:37.630 align:middle line:90% is a growing avenue of research. 00:31:37.630 --> 00:31:40.220 align:middle line:84% This gives us access to STEM cells and cells 00:31:40.220 --> 00:31:42.920 align:middle line:84% that are capable of replicating and differentiating 00:31:42.920 --> 00:31:44.930 align:middle line:84% into almost any cell in your body, 00:31:44.930 --> 00:31:48.890 align:middle line:84% and rather than dosing your entire body with a drug, 00:31:48.890 --> 00:31:52.400 align:middle line:84% you can now specifically deliver a particular cell 00:31:52.400 --> 00:31:55.560 align:middle line:84% to a particular place, and have it take its-- 00:31:55.560 --> 00:31:57.840 align:middle line:90% its role as a replacement. 00:31:57.840 --> 00:31:59.510 align:middle line:84% And so this is the kind of programming 00:31:59.510 --> 00:32:01.885 align:middle line:84% I think we should be thinking about as a grand challenge. 00:32:01.885 --> 00:32:04.940 align:middle line:84% Remember, grand challenges are not going to happen tomorrow. 00:32:04.940 --> 00:32:06.280 align:middle line:90% You have to do some piece of it. 00:32:06.280 --> 00:32:06.780 align:middle line:90% Question? 00:32:06.780 --> 00:32:09.702 align:middle line:84% AUDIENCE: In the first bullet in B, 00:32:09.702 --> 00:32:12.590 align:middle line:84% since it appears we don't really yet know 00:32:12.590 --> 00:32:14.882 align:middle line:84% what the function of a protein is going to be unless we 00:32:14.882 --> 00:32:15.862 align:middle line:90% know how it folds, right? 00:32:15.862 --> 00:32:16.737 align:middle line:90% GEORGE CHURCH: Right. 00:32:16.737 --> 00:32:19.910 align:middle line:84% AUDIENCE: So we might be able to model the sequence, 00:32:19.910 --> 00:32:22.710 align:middle line:84% but how would we model anything other than that? 00:32:22.710 --> 00:32:24.263 align:middle line:84% GEORGE CHURCH: So the question is, 00:32:24.263 --> 00:32:26.430 align:middle line:84% what do we know about-- or, rephrasing the question. 00:32:26.430 --> 00:32:28.763 align:middle line:84% What do we know about a protein before we know its fold? 00:32:28.763 --> 00:32:31.470 align:middle line:84% And actually historically we knew more about protein-- 00:32:31.470 --> 00:32:33.540 align:middle line:84% much more about protein function than we 00:32:33.540 --> 00:32:37.200 align:middle line:84% knew about their folding because the-- 00:32:37.200 --> 00:32:39.690 align:middle line:84% and we'll get to some of the definitions of functions 00:32:39.690 --> 00:32:40.720 align:middle line:90% in just a moment. 00:32:40.720 --> 00:32:42.750 align:middle line:84% But you can study it biochemically 00:32:42.750 --> 00:32:45.360 align:middle line:84% in terms of what it binds to, what 00:32:45.360 --> 00:32:48.790 align:middle line:84% place it-- it holds in the replication of the cell, 00:32:48.790 --> 00:32:50.190 align:middle line:90% and so on. 00:32:50.190 --> 00:32:53.250 align:middle line:84% We do know the folding of most proteins, 00:32:53.250 --> 00:32:56.160 align:middle line:84% and that will be-- part of the post genomic era will 00:32:56.160 --> 00:32:58.440 align:middle line:84% be producing the three dimensional 00:32:58.440 --> 00:33:02.850 align:middle line:84% structure and biochemical function of all the proteins. 00:33:02.850 --> 00:33:05.511 align:middle line:84% AUDIENCE: But here you're asking for changing that, changing 00:33:05.511 --> 00:33:06.789 align:middle line:90% the genome programming, right? 00:33:06.789 --> 00:33:07.497 align:middle line:90% And then try to-- 00:33:07.497 --> 00:33:09.983 align:middle line:84% GEORGE CHURCH: Of the parts that you understand. 00:33:09.983 --> 00:33:10.525 align:middle line:90% AUDIENCE: OK. 00:33:10.525 --> 00:33:12.160 align:middle line:84% GEORGE CHURCH: I mean, obviously we 00:33:12.160 --> 00:33:15.550 align:middle line:84% do engineer a variety of physical and biological systems 00:33:15.550 --> 00:33:18.430 align:middle line:90% without full understanding. 00:33:18.430 --> 00:33:20.950 align:middle line:84% An even grander challenge would be full understanding. 00:33:20.950 --> 00:33:24.850 align:middle line:84% Here we're trying to take subsystems that we do-- 00:33:24.850 --> 00:33:26.890 align:middle line:84% it could even be a highly integrated system 00:33:26.890 --> 00:33:28.880 align:middle line:84% where you do model the entire system, 00:33:28.880 --> 00:33:30.880 align:middle line:84% but there will always be gaps in your knowledge, 00:33:30.880 --> 00:33:33.790 align:middle line:84% just like there are gaps in the human genome sequence. 00:33:33.790 --> 00:33:36.580 align:middle line:84% And the final illustration, number C, 00:33:36.580 --> 00:33:39.955 align:middle line:84% is going up to the most complex systems 00:33:39.955 --> 00:33:46.450 align:middle line:84% that we're dealing with, which would be morphological systems 00:33:46.450 --> 00:33:47.800 align:middle line:90% and even-- 00:33:47.800 --> 00:33:50.860 align:middle line:84% and the populations that result from that. 00:33:50.860 --> 00:33:53.365 align:middle line:90% And here we will be-- 00:33:53.365 --> 00:33:55.840 align:middle line:84% you can deal with morphology in a way that-- 00:33:55.840 --> 00:33:58.090 align:middle line:84% at the molecular level all the way up 00:33:58.090 --> 00:34:02.110 align:middle line:84% through the morphology of assemblies of cells, 00:34:02.110 --> 00:34:03.550 align:middle line:90% how cells aggregate. 00:34:03.550 --> 00:34:08.230 align:middle line:84% And all this can be modeled and used to great effect, 00:34:08.230 --> 00:34:15.159 align:middle line:84% whether it's smart materials or replacements in human systems. 00:34:15.159 --> 00:34:17.409 align:middle line:84% So let's talk a little bit more about these components 00:34:17.409 --> 00:34:19.739 align:middle line:90% and how they're interconnected. 00:34:19.739 --> 00:34:22.715 align:middle line:84% Whether we are taking these components apart or putting 00:34:22.715 --> 00:34:24.090 align:middle line:84% them back together again, we need 00:34:24.090 --> 00:34:27.328 align:middle line:84% to understand how many of them there are and-- 00:34:27.328 --> 00:34:29.940 align:middle line:84% and how we're going to access them in databases. 00:34:29.940 --> 00:34:36.300 align:middle line:84% I'm illustrating this with three organisms that are nicely 00:34:36.300 --> 00:34:39.239 align:middle line:84% poised to show the extremes on the left and right 00:34:39.239 --> 00:34:40.570 align:middle line:90% and something in between. 00:34:40.570 --> 00:34:42.690 align:middle line:84% So this is mycoplasma pneumoniae-- 00:34:42.690 --> 00:34:47.770 align:middle line:84% sorry, mycoplasma genitalium, one of the smallest living 00:34:47.770 --> 00:34:50.050 align:middle line:90% organisms and smallest genome. 00:34:50.050 --> 00:34:54.820 align:middle line:84% Its genome size is a little over half a million base pairs. 00:34:54.820 --> 00:34:58.630 align:middle line:84% The worm caenorhabditis elegans, one of the first metazoan 00:34:58.630 --> 00:35:00.580 align:middle line:84% multicellular organisms sequenced, 00:35:00.580 --> 00:35:04.270 align:middle line:84% was a little less than 100 million bases, and the human 00:35:04.270 --> 00:35:05.260 align:middle line:90% at 3 billion bases. 00:35:05.260 --> 00:35:10.790 align:middle line:84% Neither the worm nor the human is completely sequenced, 00:35:10.790 --> 00:35:15.240 align:middle line:84% despite some possible indications to the contrary. 00:35:15.240 --> 00:35:18.330 align:middle line:84% There are quite a number of gaps in each. 00:35:18.330 --> 00:35:21.480 align:middle line:84% The many bacterial genomes are completely sequenced, 00:35:21.480 --> 00:35:22.412 align:middle line:90% including mycoplasma. 00:35:22.412 --> 00:35:24.120 align:middle line:84% The number of DNAs in each of these-- you 00:35:24.120 --> 00:35:29.640 align:middle line:84% have one circular genome in many bacteria. 00:35:29.640 --> 00:35:32.400 align:middle line:90% Some have multiple chromosomes. 00:35:32.400 --> 00:35:36.870 align:middle line:84% The worm has seven chromosomes and human has 25. 00:35:36.870 --> 00:35:38.963 align:middle line:84% Those of you who have studied biology or just 00:35:38.963 --> 00:35:40.380 align:middle line:84% listened in the last lecture where 00:35:40.380 --> 00:35:44.670 align:middle line:84% I said we had 23 pairs of chromosomes to segregate, you 00:35:44.670 --> 00:35:47.895 align:middle line:84% nevertheless, there are 25 different kinds of molecules. 00:35:47.895 --> 00:35:50.370 align:middle line:84% I'll leave that as an exercise for you. 00:35:50.370 --> 00:35:54.460 align:middle line:84% You can ask me in just a couple of minutes. 00:35:54.460 --> 00:35:57.120 align:middle line:84% The number of genes encoded in these DNA 00:35:57.120 --> 00:35:59.010 align:middle line:84% depends on your definition of gene. 00:35:59.010 --> 00:36:03.090 align:middle line:84% If we define it somewhat arbitrarily 00:36:03.090 --> 00:36:09.270 align:middle line:84% as a piece of inherited material that encodes one or more RNAs 00:36:09.270 --> 00:36:11.880 align:middle line:84% where those one or more RNAs share 00:36:11.880 --> 00:36:17.790 align:middle line:84% some of the same inherited material-- so in principle 00:36:17.790 --> 00:36:22.050 align:middle line:84% there's inherited material, which is not nucleic acid. 00:36:22.050 --> 00:36:25.140 align:middle line:84% But for most intents and purposes, these-- 00:36:25.140 --> 00:36:27.450 align:middle line:84% the genes that we'll be interested in 00:36:27.450 --> 00:36:31.520 align:middle line:84% do pass through RNA on their way to protein. 00:36:31.520 --> 00:36:34.070 align:middle line:84% And the number of genes in mycoplasma 00:36:34.070 --> 00:36:36.380 align:middle line:84% is roughly one gene per kilobase-- 00:36:36.380 --> 00:36:38.960 align:middle line:90% it's about 500 genes or so. 00:36:38.960 --> 00:36:45.360 align:middle line:84% The number in worms is estimated at around 20,000, 00:36:45.360 --> 00:36:50.850 align:middle line:84% and in humans it ranges from 30,000 to 150,000. 00:36:50.850 --> 00:36:54.000 align:middle line:84% And there are betting pools on exactly how many there are, 00:36:54.000 --> 00:36:56.790 align:middle line:84% and there probably should be a betting 00:36:56.790 --> 00:37:00.420 align:middle line:84% pool on when it is we will know how many there are. 00:37:00.420 --> 00:37:04.590 align:middle line:84% It's been announced at various times, but to some extent 00:37:04.590 --> 00:37:10.200 align:middle line:84% the exact number will have some softness to it 00:37:10.200 --> 00:37:16.010 align:middle line:84% because some of the genes will be of marginal utility to-- 00:37:16.010 --> 00:37:16.950 align:middle line:90% to humans. 00:37:16.950 --> 00:37:21.940 align:middle line:84% They will have been of some consequence 00:37:21.940 --> 00:37:25.920 align:middle line:84% maybe many generations ago, but on a day to day basis 00:37:25.920 --> 00:37:29.190 align:middle line:84% they will be hard to detect the importance of whether that gene 00:37:29.190 --> 00:37:30.090 align:middle line:90% is present or not. 00:37:30.090 --> 00:37:34.410 align:middle line:90% 00:37:34.410 --> 00:37:40.090 align:middle line:84% In terms of RNAs, in bacteria you 00:37:40.090 --> 00:37:44.080 align:middle line:84% have a tendency to have more genes than you 00:37:44.080 --> 00:37:49.150 align:middle line:84% have RNAs because the genes will be constructed in a series such 00:37:49.150 --> 00:37:53.290 align:middle line:84% that one RNA can make it through multiple genes in an operon, 00:37:53.290 --> 00:37:56.780 align:middle line:84% then that operon will then make multiple proteins. 00:37:56.780 --> 00:38:01.580 align:middle line:84% So you might have slightly fewer RNAs than you have genes. 00:38:01.580 --> 00:38:05.900 align:middle line:84% And worms are an example of a multicellular organism 00:38:05.900 --> 00:38:08.840 align:middle line:84% that also has operons where genes will be strung together. 00:38:08.840 --> 00:38:12.470 align:middle line:84% They tend to be shorter and fewer in number, 00:38:12.470 --> 00:38:15.747 align:middle line:84% but then they-- but that doesn't just reduce the number of RNAs. 00:38:15.747 --> 00:38:17.330 align:middle line:84% You can then increase them because you 00:38:17.330 --> 00:38:20.870 align:middle line:84% have alternative splicing where the RNAs will be made up 00:38:20.870 --> 00:38:25.220 align:middle line:84% of multiple pieces called exons which are stitched together 00:38:25.220 --> 00:38:30.590 align:middle line:84% by a specific biochemical machine-- 00:38:30.590 --> 00:38:32.120 align:middle line:90% splicing machinery. 00:38:32.120 --> 00:38:35.690 align:middle line:84% And that could happen in more than one way. 00:38:35.690 --> 00:38:38.990 align:middle line:84% They tend to be in a linear order in the genome, 00:38:38.990 --> 00:38:41.840 align:middle line:84% but there's exotic mechanisms like trans-splicing where 00:38:41.840 --> 00:38:44.330 align:middle line:84% you pull up an exon from completely 00:38:44.330 --> 00:38:47.880 align:middle line:84% different part of the genome and then splice it in together. 00:38:47.880 --> 00:38:50.480 align:middle line:84% So this number is larger than the number of genes 00:38:50.480 --> 00:38:52.735 align:middle line:84% because one gene can produce multiple RNAs 00:38:52.735 --> 00:38:53.735 align:middle line:90% by alternative splicing. 00:38:53.735 --> 00:38:58.110 align:middle line:90% 00:38:58.110 --> 00:39:01.470 align:middle line:84% For proteins, there's additional diversification 00:39:01.470 --> 00:39:04.380 align:middle line:84% that you can modify the RNAs in various ways. 00:39:04.380 --> 00:39:06.450 align:middle line:84% In prokaryotes the number of RNA modifications 00:39:06.450 --> 00:39:09.510 align:middle line:84% is relatively limited, but the number of protein modifications 00:39:09.510 --> 00:39:10.510 align:middle line:90% starts to go up. 00:39:10.510 --> 00:39:14.730 align:middle line:84% You can have proteolytic modifications, phosphorylation 00:39:14.730 --> 00:39:17.620 align:middle line:90% of various amino acids. 00:39:17.620 --> 00:39:21.070 align:middle line:84% And-- and in multicellular organisms 00:39:21.070 --> 00:39:25.000 align:middle line:84% like worms and humans, the number of modifications 00:39:25.000 --> 00:39:27.730 align:middle line:84% reaches up into about 250 different amino acids-- 00:39:27.730 --> 00:39:31.390 align:middle line:84% modified amino acids of the basic 20 amino acids, which 00:39:31.390 --> 00:39:33.910 align:middle line:84% we'll talk about in just a moment. 00:39:33.910 --> 00:39:37.780 align:middle line:84% The number of cell types in a very simple organism 00:39:37.780 --> 00:39:40.750 align:middle line:84% in a very simple environment might be as little as one. 00:39:40.750 --> 00:39:43.420 align:middle line:84% We don't really know how many cell types there are, 00:39:43.420 --> 00:39:47.130 align:middle line:84% but basically all the cell types for an organism like mycoplasma 00:39:47.130 --> 00:39:49.420 align:middle line:84% look fairly similar morphologically 00:39:49.420 --> 00:39:52.030 align:middle line:90% and probably functionally. 00:39:52.030 --> 00:39:58.720 align:middle line:84% On the other hand, the worm has 595, 500, 959 cells, 00:39:58.720 --> 00:39:59.650 align:middle line:90% and those three-- 00:39:59.650 --> 00:40:01.290 align:middle line:84% this is three significant figures. 00:40:01.290 --> 00:40:02.770 align:middle line:90% This is pretty good for biology. 00:40:02.770 --> 00:40:06.550 align:middle line:84% And these are non-gonadal cells, and the reason we know this so 00:40:06.550 --> 00:40:10.333 align:middle line:84% precisely is the entire lineage, the entire division of all 00:40:10.333 --> 00:40:12.250 align:middle line:84% the cells, have been mapped out for this worm. 00:40:12.250 --> 00:40:14.350 align:middle line:84% And we'll show this a few slides from now. 00:40:14.350 --> 00:40:18.790 align:middle line:84% Humans on the other hand, not only is the number of-- 00:40:18.790 --> 00:40:21.220 align:middle line:84% the lineages is very poorly defined 00:40:21.220 --> 00:40:24.670 align:middle line:84% for most of the cell types in the human body, 00:40:24.670 --> 00:40:27.205 align:middle line:84% and even the number of cell types is unknown. 00:40:27.205 --> 00:40:29.720 align:middle line:90% 00:40:29.720 --> 00:40:32.570 align:middle line:84% Some people will estimate as few as 200 cell types. 00:40:32.570 --> 00:40:36.040 align:middle line:84% This is just a soundbite that is made up as far as I can tell. 00:40:36.040 --> 00:40:38.950 align:middle line:90% Some people say 200,000. 00:40:38.950 --> 00:40:41.650 align:middle line:84% It's probably a safe bet that a given-- at any given time 00:40:41.650 --> 00:40:44.260 align:middle line:84% point there are fewer than 10 to the 14th cell types. 00:40:44.260 --> 00:40:45.760 align:middle line:84% This is probably not very reassuring 00:40:45.760 --> 00:40:50.110 align:middle line:84% for those of you who would like to, say, measure expression 00:40:50.110 --> 00:40:53.560 align:middle line:84% in all the different cell types, because 10 00:40:53.560 --> 00:40:57.430 align:middle line:84% to the 14th expression patterns would be quite a number. 00:40:57.430 --> 00:41:03.280 align:middle line:84% In addition, you have various developmental stages 00:41:03.280 --> 00:41:06.550 align:middle line:90% where let's say you have-- 00:41:06.550 --> 00:41:10.750 align:middle line:84% as you grow from single cell to 10 to the 14th cells, 00:41:10.750 --> 00:41:13.120 align:middle line:84% you pass through stages, and what may be-- 00:41:13.120 --> 00:41:17.110 align:middle line:84% appear to be the same cell type at one stage at an earlier time 00:41:17.110 --> 00:41:19.910 align:middle line:84% point may have completely different gene expression. 00:41:19.910 --> 00:41:23.390 align:middle line:84% That is known, even though the total number of cell types 00:41:23.390 --> 00:41:23.890 align:middle line:90% is not. 00:41:23.890 --> 00:41:27.016 align:middle line:84% AUDIENCE: [INAUDIBLE] the RNA, do you know how many 00:41:27.016 --> 00:41:30.824 align:middle line:84% [? extra ?] [INAUDIBLE] to [? include ?] RNA [INAUDIBLE]?? 00:41:30.824 --> 00:41:32.210 align:middle line:90% GEORGE CHURCH: OK, yes. 00:41:32.210 --> 00:41:37.610 align:middle line:84% I meant to caution you that-- just that the terminology here 00:41:37.610 --> 00:41:40.880 align:middle line:90% is used quite loosely. 00:41:40.880 --> 00:41:43.220 align:middle line:84% Gene expression is often used interchangeably 00:41:43.220 --> 00:41:46.190 align:middle line:90% with RNA expression, clearly. 00:41:46.190 --> 00:41:48.860 align:middle line:84% And then in almost the same paper, 00:41:48.860 --> 00:41:51.860 align:middle line:84% they will refer to as genes-- as protein encoding genes, 00:41:51.860 --> 00:41:56.930 align:middle line:84% completely sidestepping a large number of RNAs which really 00:41:56.930 --> 00:42:00.740 align:middle line:84% are never translated into proteins, 00:42:00.740 --> 00:42:07.680 align:middle line:84% such as ribosomal RNAs, tRNAs, small nuclear RNAs, 00:42:07.680 --> 00:42:10.018 align:middle line:84% and a whole variety of regulatory RNAs, RNAis 00:42:10.018 --> 00:42:10.560 align:middle line:90% and so forth. 00:42:10.560 --> 00:42:13.140 align:middle line:84% It's becoming very important, this class of RNAs 00:42:13.140 --> 00:42:17.190 align:middle line:84% which stay as RNAs, so be careful when 00:42:17.190 --> 00:42:19.740 align:middle line:84% people use genes and RNAs interchangeably, 00:42:19.740 --> 00:42:21.750 align:middle line:84% or genes and proteins interchangeably. 00:42:21.750 --> 00:42:26.550 align:middle line:90% 00:42:26.550 --> 00:42:31.020 align:middle line:84% So this is an example of molecular morphology, 00:42:31.020 --> 00:42:37.470 align:middle line:84% a particularly elegant example that illustrated last time. 00:42:37.470 --> 00:42:39.750 align:middle line:84% In a certain sense, the morphology of these two 00:42:39.750 --> 00:42:42.600 align:middle line:90% strands of DNA greatly-- 00:42:42.600 --> 00:42:48.270 align:middle line:84% go a long way towards explaining the inheritance and fidelity 00:42:48.270 --> 00:42:53.220 align:middle line:84% of the basic macromolecule which stores the information. 00:42:53.220 --> 00:42:56.600 align:middle line:84% What we're going to do is expand on the-- 00:42:56.600 --> 00:43:00.050 align:middle line:84% look outside of these bases which form the base 00:43:00.050 --> 00:43:03.440 align:middle line:84% pairs down the-- that has stacked up 00:43:03.440 --> 00:43:06.500 align:middle line:84% along the core of the DNA to look at how they're actually 00:43:06.500 --> 00:43:09.320 align:middle line:84% covalently attached, what the precursors are when we 00:43:09.320 --> 00:43:10.798 align:middle line:90% go from monomers to polymers. 00:43:10.798 --> 00:43:12.590 align:middle line:84% We're going to talk about polymer synthesis 00:43:12.590 --> 00:43:14.940 align:middle line:90% for the next few slides. 00:43:14.940 --> 00:43:19.740 align:middle line:84% So in order to get this-- this exquisite base pairing here, 00:43:19.740 --> 00:43:22.790 align:middle line:84% which a recent article has argued is optimal. 00:43:22.790 --> 00:43:24.290 align:middle line:84% Of all the different base pairs that 00:43:24.290 --> 00:43:29.150 align:middle line:84% could have formed in the prebiotic times, 00:43:29.150 --> 00:43:32.660 align:middle line:84% these are some of the optimal alignment of hydrogen bonds. 00:43:32.660 --> 00:43:34.880 align:middle line:84% But the hydrogen bonds just guides the base pairing, 00:43:34.880 --> 00:43:37.850 align:middle line:84% the polymerization occurs in something not 00:43:37.850 --> 00:43:41.810 align:middle line:84% shown on this slide, but shown on the next slide. 00:43:41.810 --> 00:43:47.810 align:middle line:84% Here are the two examples, two very similar bases 00:43:47.810 --> 00:43:51.800 align:middle line:84% for DNA and RNA, the monomers that 00:43:51.800 --> 00:43:55.010 align:middle line:90% are polymerized by enzymes-- 00:43:55.010 --> 00:43:58.830 align:middle line:84% polymerases-- to make DNA and RNA. 00:43:58.830 --> 00:44:03.500 align:middle line:84% So on the top is the deoxy-ATP and below is the ribo-ATP. 00:44:03.500 --> 00:44:06.140 align:middle line:84% Ribo-ATP is distinguished not only as 00:44:06.140 --> 00:44:07.790 align:middle line:84% a precursor for polymer, but it's 00:44:07.790 --> 00:44:13.070 align:middle line:84% one of the key biomolecules providing energy 00:44:13.070 --> 00:44:15.620 align:middle line:84% or transmitting energy from one part of the cell 00:44:15.620 --> 00:44:17.390 align:middle line:84% to another, one machine to another. 00:44:17.390 --> 00:44:21.950 align:middle line:84% If you look at a network diagram of the metabolism, 00:44:21.950 --> 00:44:25.100 align:middle line:84% ribo-ATP would probably be one of the most-- 00:44:25.100 --> 00:44:29.550 align:middle line:84% is one of the most highly connected nodes in that graph. 00:44:29.550 --> 00:44:33.800 align:middle line:84% It's connected hundreds of times in a graph where many things 00:44:33.800 --> 00:44:36.630 align:middle line:90% are connected once or twice. 00:44:36.630 --> 00:44:39.630 align:middle line:84% And this is the structure, this is the base here 00:44:39.630 --> 00:44:41.790 align:middle line:84% in skeletal form and space filling form. 00:44:41.790 --> 00:44:45.630 align:middle line:84% The space filling form, aside from the colors, is-- 00:44:45.630 --> 00:44:48.150 align:middle line:84% is getting to be a more accurate representation 00:44:48.150 --> 00:44:49.747 align:middle line:90% of the electron density-- 00:44:49.747 --> 00:44:51.330 align:middle line:84% sort of the electron density you might 00:44:51.330 --> 00:44:54.780 align:middle line:84% observe in crystallography or in quantum calculations. 00:44:54.780 --> 00:44:56.695 align:middle line:84% But the skeletal form allows you to see 00:44:56.695 --> 00:44:58.570 align:middle line:84% some of the hidden atoms a little bit better. 00:44:58.570 --> 00:45:01.740 align:middle line:84% You can see the nitrogens are color coded for-- 00:45:01.740 --> 00:45:05.280 align:middle line:84% by blue, and the phosphates and sulfurs-- 00:45:05.280 --> 00:45:08.730 align:middle line:84% in this case, phosphates by yellow and the oxygens by red. 00:45:08.730 --> 00:45:11.770 align:middle line:90% Carbons gray or black. 00:45:11.770 --> 00:45:13.990 align:middle line:84% And so what you see is the only real difference 00:45:13.990 --> 00:45:16.660 align:middle line:84% between the deoxy-ATP and the ribo-ATP 00:45:16.660 --> 00:45:21.080 align:middle line:84% refers to this oxygen at the two prime position, 00:45:21.080 --> 00:45:23.240 align:middle line:90% which is the deoxy. 00:45:23.240 --> 00:45:27.850 align:middle line:90% 00:45:27.850 --> 00:45:32.362 align:middle line:90% They-- they both share the-- 00:45:32.362 --> 00:45:34.835 align:middle line:84% this ribose and phosphate which were 00:45:34.835 --> 00:45:37.255 align:middle line:84% what was not represented on the previous slide 00:45:37.255 --> 00:45:38.860 align:middle line:90% are the repeating backbone. 00:45:38.860 --> 00:45:41.337 align:middle line:84% You go from this 3 prime hydroxyl-- 00:45:41.337 --> 00:45:43.420 align:middle line:84% the numbering here, by the way, for the ribose has 00:45:43.420 --> 00:45:44.920 align:middle line:84% primes after it to distinguish it 00:45:44.920 --> 00:45:46.360 align:middle line:90% from the numbering of the basis. 00:45:46.360 --> 00:45:49.990 align:middle line:84% These were studied chemically by chemists 00:45:49.990 --> 00:45:51.790 align:middle line:84% and numbered independently, and so when 00:45:51.790 --> 00:45:53.410 align:middle line:84% they were found in the same molecule, 00:45:53.410 --> 00:45:54.535 align:middle line:90% you had to have a separate. 00:45:54.535 --> 00:45:56.920 align:middle line:84% So that's the reason, throughout the rest of this course 00:45:56.920 --> 00:45:58.420 align:middle line:84% and the rest of your life probably, 00:45:58.420 --> 00:46:01.390 align:middle line:84% you'll be referring to things going from 5 prime to 3 prime. 00:46:01.390 --> 00:46:05.500 align:middle line:84% It's because the people studying the bases won over the people 00:46:05.500 --> 00:46:08.160 align:middle line:90% studying the riboses. 00:46:08.160 --> 00:46:10.680 align:middle line:84% Anyway, so the last two phosphates just 00:46:10.680 --> 00:46:15.030 align:middle line:84% provide a higher energy bonds, and [? these ?] by equilibrium 00:46:15.030 --> 00:46:21.300 align:middle line:84% is pushed so that this whole splitting in polymerization 00:46:21.300 --> 00:46:25.625 align:middle line:84% is a very favorable in terms of free energy. 00:46:25.625 --> 00:46:26.125 align:middle line:90% OK. 00:46:26.125 --> 00:46:28.720 align:middle line:90% 00:46:28.720 --> 00:46:32.770 align:middle line:84% Now as we discussed-- so those were 00:46:32.770 --> 00:46:36.385 align:middle line:84% the nucleic acid components, and then the proteins 00:46:36.385 --> 00:46:37.760 align:middle line:84% that they encode and the proteins 00:46:37.760 --> 00:46:39.302 align:middle line:84% that are required for the replication 00:46:39.302 --> 00:46:41.260 align:middle line:84% of the nucleic acid components are made up 00:46:41.260 --> 00:46:43.210 align:middle line:84% of simple derivatives of glycine, 00:46:43.210 --> 00:46:46.240 align:middle line:84% which can be represented-- full name, three letter code, 00:46:46.240 --> 00:46:47.530 align:middle line:90% one letter code. 00:46:47.530 --> 00:46:50.410 align:middle line:84% You should learn the 20 one letter codes 00:46:50.410 --> 00:46:52.360 align:middle line:84% because they're very valuable in this course 00:46:52.360 --> 00:46:55.683 align:middle line:84% and in bioinformatics in general. 00:46:55.683 --> 00:46:57.100 align:middle line:84% Again, the same color coding here. 00:46:57.100 --> 00:46:58.420 align:middle line:90% You have nitrogen car-- 00:46:58.420 --> 00:47:00.520 align:middle line:84% this is the central carbon, it's the alpha, 00:47:00.520 --> 00:47:02.017 align:middle line:90% and this is a carboxyl group. 00:47:02.017 --> 00:47:03.850 align:middle line:84% So this is-- as an amino [? acid, this ?] is 00:47:03.850 --> 00:47:06.017 align:middle line:84% a [INAUDIBLE] ion with a positively charged nitrogen 00:47:06.017 --> 00:47:07.870 align:middle line:84% and a negatively charged carboxylate. 00:47:07.870 --> 00:47:10.750 align:middle line:84% And the way you represent this in a computer, 00:47:10.750 --> 00:47:14.020 align:middle line:84% you can either represent it as a pretty picture here, 00:47:14.020 --> 00:47:16.720 align:middle line:84% either skeletal or space filling. 00:47:16.720 --> 00:47:19.250 align:middle line:84% That's of course represented by zeros and ones, 00:47:19.250 --> 00:47:22.060 align:middle line:84% but it's not a very useful way for searching, merging, 00:47:22.060 --> 00:47:22.930 align:middle line:90% or checking, right? 00:47:22.930 --> 00:47:24.938 align:middle line:84% It's going to-- if you rotate this slightly 00:47:24.938 --> 00:47:26.980 align:middle line:84% in three dimensions, it's going to [? give ?] you 00:47:26.980 --> 00:47:30.490 align:middle line:84% a completely different image and it'll be hard to search. 00:47:30.490 --> 00:47:33.370 align:middle line:84% You could represent it as the three-- 00:47:33.370 --> 00:47:35.590 align:middle line:84% the coordinates, the x, y, and z coordinates 00:47:35.590 --> 00:47:37.840 align:middle line:90% of each of these atoms. 00:47:37.840 --> 00:47:39.940 align:middle line:84% And that is something that you can search, 00:47:39.940 --> 00:47:44.020 align:middle line:84% but you also need to represent it in a-- in a way that 00:47:44.020 --> 00:47:49.030 align:middle line:84% represents the hierarchical structure by which these things 00:47:49.030 --> 00:47:53.530 align:middle line:84% form covalent bonds, and so that you can recognize groups-- 00:47:53.530 --> 00:47:57.370 align:middle line:84% groupings of atoms into polymers, 00:47:57.370 --> 00:48:00.790 align:middle line:84% polymers into assemblies, and so all the way up. 00:48:00.790 --> 00:48:04.300 align:middle line:84% And this is an example of such a hierarchical description which 00:48:04.300 --> 00:48:09.160 align:middle line:84% would be recognizable to all the computer scientists 00:48:09.160 --> 00:48:12.250 align:middle line:84% here if they were comfortable with some 00:48:12.250 --> 00:48:14.210 align:middle line:90% of the biochemical terms. 00:48:14.210 --> 00:48:16.643 align:middle line:84% So here's the configuration of this thing 00:48:16.643 --> 00:48:17.560 align:middle line:90% we're calling glycine. 00:48:17.560 --> 00:48:20.470 align:middle line:90% 00:48:20.470 --> 00:48:22.360 align:middle line:84% By the way, there'll be a lot of jargon 00:48:22.360 --> 00:48:23.860 align:middle line:84% in this [? court ?] for those of you 00:48:23.860 --> 00:48:26.720 align:middle line:90% that are computer scientists. 00:48:26.720 --> 00:48:29.090 align:middle line:84% The point of this course is not to give you 00:48:29.090 --> 00:48:31.460 align:middle line:84% an encyclopedic knowledge, it's more to flesh 00:48:31.460 --> 00:48:33.710 align:middle line:84% out the concepts [? that ?] supplies both the computer 00:48:33.710 --> 00:48:35.660 align:middle line:90% and biologists in the group. 00:48:35.660 --> 00:48:37.430 align:middle line:84% And so if you learn a lot of facts, 00:48:37.430 --> 00:48:39.740 align:middle line:84% then we'll hold it against you, but-- 00:48:39.740 --> 00:48:41.640 align:middle line:84% but every time you see a piece of jargon 00:48:41.640 --> 00:48:43.098 align:middle line:84% and you think I haven't defined it, 00:48:43.098 --> 00:48:47.910 align:middle line:84% just call it Fred or George or something like that. 00:48:47.910 --> 00:48:49.190 align:middle line:90% It's an arbitrary name. 00:48:49.190 --> 00:48:51.140 align:middle line:84% It will be defined in the databases, 00:48:51.140 --> 00:48:53.060 align:middle line:84% and that's what the database is for-- keeping tract of this. 00:48:53.060 --> 00:48:54.590 align:middle line:84% But you will have to understand the concepts, 00:48:54.590 --> 00:48:55.760 align:middle line:84% and the concept here we're trying 00:48:55.760 --> 00:48:58.040 align:middle line:84% to illustrate [? are ?] different ways of representing 00:48:58.040 --> 00:49:04.490 align:middle line:84% the molecular definitions, here by describing the-- the syntax 00:49:04.490 --> 00:49:08.960 align:middle line:84% as you would in breaking up a English sentence 00:49:08.960 --> 00:49:10.890 align:middle line:90% into its structure. 00:49:10.890 --> 00:49:11.900 align:middle line:90% So you have here-- 00:49:11.900 --> 00:49:14.570 align:middle line:84% it has a substituent of a backbone. 00:49:14.570 --> 00:49:18.717 align:middle line:84% Here in order to try to tie together all amino acids, 00:49:18.717 --> 00:49:21.050 align:middle line:84% you've made something that's a little nonsensical, which 00:49:21.050 --> 00:49:24.650 align:middle line:84% is you talk about the L backbone of amino acid. 00:49:24.650 --> 00:49:26.750 align:middle line:84% All the amino acids except for glycine 00:49:26.750 --> 00:49:28.377 align:middle line:90% actually have a handedness. 00:49:28.377 --> 00:49:30.710 align:middle line:84% That is to say, if you hold them up in a mirror it looks 00:49:30.710 --> 00:49:32.880 align:middle line:84% different from what the thing that you're holding your hand 00:49:32.880 --> 00:49:34.550 align:middle line:84% if you take, say, a space filling [? model and hold ?] 00:49:34.550 --> 00:49:35.550 align:middle line:90% [? it up to a ?] mirror. 00:49:35.550 --> 00:49:37.550 align:middle line:84% And the reason is that you can have-- 00:49:37.550 --> 00:49:39.680 align:middle line:84% these two hydrogens here in glycine 00:49:39.680 --> 00:49:43.640 align:middle line:84% will have an actual side chain coming off in amino acids. 00:49:43.640 --> 00:49:48.440 align:middle line:84% And if it comes off here, then it's [? a D ?] amino acid. 00:49:48.440 --> 00:49:52.110 align:middle line:84% If it comes off of the other hydrogen, it's a L amino acid. 00:49:52.110 --> 00:49:55.140 align:middle line:84% So here you're saying natural amino acids are L amino acids, 00:49:55.140 --> 00:49:56.870 align:middle line:84% and so you want to take this L backbone 00:49:56.870 --> 00:49:58.130 align:middle line:90% and put a substituent on it. 00:49:58.130 --> 00:50:01.980 align:middle line:84% That substituent is HYD for hydrogen, 00:50:01.980 --> 00:50:07.380 align:middle line:84% and it's linked through carbon [? 1 ?] to another hydrogen 00:50:07.380 --> 00:50:08.250 align:middle line:90% and so forth. 00:50:08.250 --> 00:50:09.510 align:middle line:90% Nil means nothing. 00:50:09.510 --> 00:50:12.330 align:middle line:84% And here's another way of representing it slightly more 00:50:12.330 --> 00:50:12.900 align:middle line:90% compact. 00:50:12.900 --> 00:50:14.640 align:middle line:84% You can think of this as just one long string 00:50:14.640 --> 00:50:16.098 align:middle line:84% even though it's on multiple lines. 00:50:16.098 --> 00:50:18.880 align:middle line:84% Here's one that's definitely [? at ?] the bottom line 00:50:18.880 --> 00:50:23.280 align:middle line:84% [? is ?] you've got this CH2 group, this methylene group, 00:50:23.280 --> 00:50:27.090 align:middle line:84% right in the middle bounded by this positively charged 00:50:27.090 --> 00:50:30.110 align:middle line:84% nitrogen and negatively charged carboxyl group. 00:50:30.110 --> 00:50:33.730 align:middle line:84% So you can see these nested, parenthetical ways 00:50:33.730 --> 00:50:35.740 align:middle line:90% of indicating the hierarchy. 00:50:35.740 --> 00:50:38.590 align:middle line:84% This allow you to search through complicated databases 00:50:38.590 --> 00:50:41.440 align:middle line:84% of compounds looking for shared properties, 00:50:41.440 --> 00:50:44.320 align:middle line:84% say, of all the drugs that bind to a receptor whether 00:50:44.320 --> 00:50:45.875 align:middle line:84% or not the structure of the receptor. 00:50:45.875 --> 00:50:47.500 align:middle line:84% If you know the structure of the drugs, 00:50:47.500 --> 00:50:51.500 align:middle line:84% you can do a structure activity relationship. 00:50:51.500 --> 00:50:52.292 align:middle line:90% AUDIENCE: Question. 00:50:52.292 --> 00:50:53.125 align:middle line:90% GEORGE CHURCH: Yeah? 00:50:53.125 --> 00:50:54.560 align:middle line:84% AUDIENCE: What's the significance 00:50:54.560 --> 00:50:58.800 align:middle line:84% of the fact that you've got the amino group 00:50:58.800 --> 00:51:02.282 align:middle line:84% and the carboxyl group both at the end, 00:51:02.282 --> 00:51:10.012 align:middle line:84% and what is usually in the middle is brought up 00:51:10.012 --> 00:51:12.910 align:middle line:90% to the left? 00:51:12.910 --> 00:51:15.740 align:middle line:84% GEORGE CHURCH: You're asking why [? in ?] a particular order? 00:51:15.740 --> 00:51:16.700 align:middle line:84% Well you can think of this [? likely ?] 00:51:16.700 --> 00:51:17.617 align:middle line:90% that your calculators. 00:51:17.617 --> 00:51:20.690 align:middle line:84% You can either enter them in the natural way that you do it, 00:51:20.690 --> 00:51:23.255 align:middle line:84% or you can do reverse [? Polish ?] notation. 00:51:23.255 --> 00:51:26.427 align:middle line:84% And different-- different ways of setting up syntax 00:51:26.427 --> 00:51:27.510 align:middle line:90% have this different thing. 00:51:27.510 --> 00:51:29.093 align:middle line:84% If you really wanted to research this, 00:51:29.093 --> 00:51:31.250 align:middle line:84% you'd look into the SMILES definition. 00:51:31.250 --> 00:51:33.530 align:middle line:84% This is a particular chemical definition, 00:51:33.530 --> 00:51:36.560 align:middle line:84% and they could justify this much better than I could. 00:51:36.560 --> 00:51:39.080 align:middle line:90% 00:51:39.080 --> 00:51:44.820 align:middle line:84% So there are 20 amino acids in the simple genetic code. 00:51:44.820 --> 00:51:48.810 align:middle line:84% There are 280 that are post-synthetic modifications 00:51:48.810 --> 00:51:50.830 align:middle line:90% of these simple 20 amino acids. 00:51:50.830 --> 00:51:53.850 align:middle line:84% We've been talking about glycine as the basic backbone shown 00:51:53.850 --> 00:51:58.350 align:middle line:84% here in black on this slide, and in blue 00:51:58.350 --> 00:52:02.361 align:middle line:84% are these side chains that lended its-- 00:52:02.361 --> 00:52:06.450 align:middle line:84% its chiral nature, its nature that has a mirror image. 00:52:06.450 --> 00:52:08.880 align:middle line:84% And each of them provides the properties 00:52:08.880 --> 00:52:10.170 align:middle line:90% which are color coded here. 00:52:10.170 --> 00:52:13.440 align:middle line:84% Orange have the property that the side chains, and hence 00:52:13.440 --> 00:52:16.560 align:middle line:84% the amino acid and the protein, are hydrophobic. 00:52:16.560 --> 00:52:19.950 align:middle line:84% They try to get out of water, they are not-- 00:52:19.950 --> 00:52:23.670 align:middle line:84% they try to bury themselves in other hydrophobic moieties, 00:52:23.670 --> 00:52:26.340 align:middle line:84% like other amino acids like this in the core of proteins 00:52:26.340 --> 00:52:30.690 align:middle line:84% or in lipids which are hydrophobic as well. 00:52:30.690 --> 00:52:32.430 align:middle line:90% Green are hydrophilic. 00:52:32.430 --> 00:52:34.227 align:middle line:84% Blue and red are also hydrophilic, 00:52:34.227 --> 00:52:35.310 align:middle line:90% but they're not only that. 00:52:35.310 --> 00:52:38.580 align:middle line:84% So they're also charged, the red ones being negatively charged 00:52:38.580 --> 00:52:41.820 align:middle line:84% as in the red of oxygen and blue being positively charged 00:52:41.820 --> 00:52:45.240 align:middle line:84% as in the blue of nitrogen. And the yellow being 00:52:45.240 --> 00:52:48.480 align:middle line:84% the sulfur containing, moderately hydrophilic 00:52:48.480 --> 00:52:49.870 align:middle line:90% amino acids. 00:52:49.870 --> 00:52:52.950 align:middle line:84% You can have more than one chiral center of symmetry, 00:52:52.950 --> 00:52:54.350 align:middle line:90% like this mirror image. 00:52:54.350 --> 00:52:59.330 align:middle line:84% You can have two such centers as in three [? inning. ?] 00:52:59.330 --> 00:53:01.130 align:middle line:84% So now we're going to put these amino acids 00:53:01.130 --> 00:53:05.320 align:middle line:90% in the context of a-- 00:53:05.320 --> 00:53:08.620 align:middle line:84% the central dogma, going from DNA to RNA to protein. 00:53:08.620 --> 00:53:14.420 align:middle line:84% We want to illustrate this in a case 00:53:14.420 --> 00:53:18.770 align:middle line:84% of a very complicated machine and a very elegant and simple 00:53:18.770 --> 00:53:19.620 align:middle line:90% algorithm. 00:53:19.620 --> 00:53:23.720 align:middle line:84% This algorithm is simple because biology cooperates largely 00:53:23.720 --> 00:53:24.230 align:middle line:90% with us. 00:53:24.230 --> 00:53:27.770 align:middle line:84% The code, unlike many codes in biology, 00:53:27.770 --> 00:53:32.870 align:middle line:84% is fairly universal, found in almost all organisms 00:53:32.870 --> 00:53:34.100 align:middle line:90% in the same form. 00:53:34.100 --> 00:53:38.810 align:middle line:84% And it's very strict, and with relatively few exceptions 00:53:38.810 --> 00:53:44.010 align:middle line:84% where you have three nucleotides which encode one amino acid. 00:53:44.010 --> 00:53:48.320 align:middle line:84% So there are 64 possible trinucleotides, 00:53:48.320 --> 00:53:52.670 align:middle line:84% 4 to the third power, and most of those trinucleotides 00:53:52.670 --> 00:53:54.320 align:middle line:90% encode some amino acid. 00:53:54.320 --> 00:53:56.480 align:middle line:84% The exceptions are stop codons indicated here, 00:53:56.480 --> 00:53:58.650 align:middle line:90% the little dashes in this table. 00:53:58.650 --> 00:54:00.620 align:middle line:84% And so let's just go through the table 00:54:00.620 --> 00:54:03.230 align:middle line:84% because that's the algorithm part of it. 00:54:03.230 --> 00:54:08.860 align:middle line:84% We have the color coded amino acids in here 00:54:08.860 --> 00:54:10.110 align:middle line:90% in a single [? little ?] code. 00:54:10.110 --> 00:54:13.500 align:middle line:84% Remember orange is hydrophobic, green is hydrophilic, 00:54:13.500 --> 00:54:18.690 align:middle line:84% blue is positive, and red is negative amino acids. 00:54:18.690 --> 00:54:22.620 align:middle line:84% And what we have [? as ?] an example would be AUG 00:54:22.620 --> 00:54:26.940 align:middle line:84% on the messenger RNA-- this is going from DNA to RNA-- 00:54:26.940 --> 00:54:32.520 align:middle line:84% is decoded by a complementary trinucleotide on this transfer 00:54:32.520 --> 00:54:33.060 align:middle line:90% RNA. 00:54:33.060 --> 00:54:38.300 align:middle line:84% Here it's unfolded, but in reality and in-- 00:54:38.300 --> 00:54:41.980 align:middle line:84% in last lecture and in the next couple of slides, 00:54:41.980 --> 00:54:43.320 align:middle line:90% you will see it more folded up. 00:54:43.320 --> 00:54:46.470 align:middle line:84% But this is unfolding it to show the 76 nucleotides 00:54:46.470 --> 00:54:50.835 align:middle line:84% or so of the transfer RNA, which has been preloaded 00:54:50.835 --> 00:54:55.050 align:middle line:84% with an enzyme which is truly the miraculous part 00:54:55.050 --> 00:54:57.270 align:middle line:84% of the geniculate code, which is the aminoacyl tRNA 00:54:57.270 --> 00:54:59.730 align:middle line:84% synthetases which recognize the transfer RNA, 00:54:59.730 --> 00:55:03.510 align:middle line:84% recognize this methionine away from all the other 20 00:55:03.510 --> 00:55:06.360 align:middle line:84% many acids, and put it on the right transfer RNA. 00:55:06.360 --> 00:55:08.910 align:middle line:84% Once that's done, then the rest is base pairing. 00:55:08.910 --> 00:55:11.850 align:middle line:84% Mostly very Watson-Crick base pairing or something 00:55:11.850 --> 00:55:14.670 align:middle line:84% like it where the first two positions dominate 00:55:14.670 --> 00:55:15.990 align:middle line:90% and the second one can wobble. 00:55:15.990 --> 00:55:22.020 align:middle line:84% Here a G and a U is not part of the ATGC cannon of Watson 00:55:22.020 --> 00:55:26.490 align:middle line:84% and Crick, but it's close enough and it allows some ambiguity 00:55:26.490 --> 00:55:27.520 align:middle line:90% at this third position. 00:55:27.520 --> 00:55:30.480 align:middle line:84% So for example, UUU not only encodes phenylalanine, 00:55:30.480 --> 00:55:33.540 align:middle line:90% but UUC can also-- 00:55:33.540 --> 00:55:36.300 align:middle line:84% here the triplet in this table is 00:55:36.300 --> 00:55:41.713 align:middle line:84% UXU where the X is U, C, A, or G along the top of the table. 00:55:41.713 --> 00:55:43.380 align:middle line:84% And so you can look this-- you basically 00:55:43.380 --> 00:55:46.620 align:middle line:84% look up the trinucleotide in a table like this in the computer 00:55:46.620 --> 00:55:49.208 align:middle line:84% and you can find the corresponding amino acid. 00:55:49.208 --> 00:55:51.750 align:middle line:84% This allows you to go basically from a DNA sequence to an RNA 00:55:51.750 --> 00:55:54.165 align:middle line:84% sequence to a protein in the computer. 00:55:54.165 --> 00:55:57.830 align:middle line:90% 00:55:57.830 --> 00:56:00.475 align:middle line:84% And what's-- so OK, that sounded too simple. 00:56:00.475 --> 00:56:02.600 align:middle line:84% Well, I'm going to give you a couple of slides that 00:56:02.600 --> 00:56:04.455 align:middle line:84% illustrate why it's more complicated. 00:56:04.455 --> 00:56:06.080 align:middle line:84% First of all, why it's more complicated 00:56:06.080 --> 00:56:12.190 align:middle line:84% biochemically is not only do you have that amazing protein 00:56:12.190 --> 00:56:15.670 align:middle line:84% molecule, sometimes one or two subunit proteins 00:56:15.670 --> 00:56:18.070 align:middle line:84% are sufficient to take these tRNAs here encoded 00:56:18.070 --> 00:56:21.682 align:middle line:84% by red and green where one amino acid 00:56:21.682 --> 00:56:23.140 align:middle line:84% is going to be added to the growing 00:56:23.140 --> 00:56:26.440 align:middle line:84% peptide chain on the other, and the two business 00:56:26.440 --> 00:56:30.100 align:middle line:84% ends of the molecule that are responsible for handing off 00:56:30.100 --> 00:56:31.990 align:middle line:84% amino acids where they get coupled together 00:56:31.990 --> 00:56:34.750 align:middle line:90% in this polymerization reaction. 00:56:34.750 --> 00:56:40.510 align:middle line:84% These two transfer RNAs have been properly charged 00:56:40.510 --> 00:56:44.050 align:middle line:84% by the amino [INAUDIBLE],, but then they 00:56:44.050 --> 00:56:46.870 align:middle line:84% require this truly huge apparatus, 00:56:46.870 --> 00:56:50.800 align:middle line:84% one of the largest molecular machines in the cell, arguably 00:56:50.800 --> 00:56:55.480 align:middle line:84% similar or larger than any other one, which-- which allows 00:56:55.480 --> 00:56:58.120 align:middle line:84% the messenger RNA not shown to bind to these two 00:56:58.120 --> 00:57:00.040 align:middle line:90% trinucleotides in the tRNAs. 00:57:00.040 --> 00:57:04.840 align:middle line:84% And then a catalytic reaction occurs where the amino acid 00:57:04.840 --> 00:57:08.080 align:middle line:84% is shifted from one tRNA to the other, 00:57:08.080 --> 00:57:09.490 align:middle line:90% making a growing peptide chain. 00:57:09.490 --> 00:57:11.050 align:middle line:84% And this chemical reaction here shown 00:57:11.050 --> 00:57:16.870 align:middle line:84% with all these circles and arrows for the chemical bond 00:57:16.870 --> 00:57:19.900 align:middle line:84% reformation is actually catalyzed by an RNA. 00:57:19.900 --> 00:57:22.283 align:middle line:84% This is the second RNA catalyst that we've 00:57:22.283 --> 00:57:23.450 align:middle line:90% talked about in this course. 00:57:23.450 --> 00:57:25.120 align:middle line:84% The first one was briefly mentioned 00:57:25.120 --> 00:57:29.290 align:middle line:84% on the subject of replication in last class, where 00:57:29.290 --> 00:57:33.400 align:middle line:84% you can find RNAs that can be engineered 00:57:33.400 --> 00:57:37.870 align:middle line:84% and probably existed in other scenarios which 00:57:37.870 --> 00:57:42.200 align:middle line:84% can replicate using small molecule precursors. 00:57:42.200 --> 00:57:46.090 align:middle line:84% So by far, most catalysts that we will be dealing with 00:57:46.090 --> 00:57:47.360 align:middle line:90% will be proteins. 00:57:47.360 --> 00:57:49.750 align:middle line:84% But in order to get the proteins we need this really 00:57:49.750 --> 00:57:51.310 align:middle line:90% complicated ribosime-- 00:57:51.310 --> 00:57:54.190 align:middle line:90% RNA enzyme to catalyze this. 00:57:54.190 --> 00:58:00.330 align:middle line:84% Here the white are the base pairs, the Watson Crick 00:58:00.330 --> 00:58:03.150 align:middle line:84% and non-Watson Crick base pairs of the RNA, 00:58:03.150 --> 00:58:07.500 align:middle line:84% the gold is the backbone ribose and phosphates 00:58:07.500 --> 00:58:10.620 align:middle line:84% of all of those RNAs, and then the blue 00:58:10.620 --> 00:58:12.420 align:middle line:84% are the proteins, which you can see 00:58:12.420 --> 00:58:15.990 align:middle line:84% are mostly out of the periphery, not 00:58:15.990 --> 00:58:19.530 align:middle line:84% involved in either the enzymatic reaction or the recognition 00:58:19.530 --> 00:58:23.550 align:middle line:84% reaction which does the decoding at the trinucleotide codon 00:58:23.550 --> 00:58:25.040 align:middle line:90% level. 00:58:25.040 --> 00:58:28.670 align:middle line:84% This is made up of total of three RNAs. 00:58:28.670 --> 00:58:31.135 align:middle line:84% Here you see two of the RNAs in the large subunit. 00:58:31.135 --> 00:58:32.510 align:middle line:84% There's a small subunit that fits 00:58:32.510 --> 00:58:34.580 align:middle line:84% on top of this which would mask the reaction we 00:58:34.580 --> 00:58:36.350 align:middle line:90% were interested in. 00:58:36.350 --> 00:58:39.440 align:middle line:84% Over 50 different proteins and the complete three 00:58:39.440 --> 00:58:42.740 align:middle line:84% dimensional structure is known from a variety 00:58:42.740 --> 00:58:45.570 align:middle line:90% of different organisms now. 00:58:45.570 --> 00:58:46.140 align:middle line:90% OK. 00:58:46.140 --> 00:58:52.200 align:middle line:84% Now after the break, we will then take this table that we 00:58:52.200 --> 00:58:56.970 align:middle line:84% had, and the complex biochemical machinery we [? had, ?] 00:58:56.970 --> 00:59:03.525 align:middle line:84% and turn it into a program that does the central dogma from DNA 00:59:03.525 --> 00:59:04.275 align:middle line:90% to RNA to protein. 00:59:04.275 --> 00:59:08.720 align:middle line:84% So take a brief break and come back and we'll talk about this. 00:59:08.720 --> 00:59:10.000 align:middle line:90%