The word "genetics" has come to mean the study of inheritance in living organisms, how (by what mechanisms) the parent or parents transmit certain traits to its or their offspring. Since the beginnings of Homo Sapiens it must have been fairly clear to us that: (1) people breed people, horses horses, an oak acorn will grow into an oak and so on, with very few exceptions which were called "monsters," and (2) children tend to resemble their parents, sometimes the mother more, sometimes the father. In the Neolithic, with the development of agriculture and animal husbandry, people learned the art of hybridization or cross-breeding, whereby new and more convenient kinds of domestic animals, fruits and cereals were created. Practically all food we find in the market today is the product of many hybridization processes, some very old, some new. But for the longest time this knowledge remained sporadic and empirical, without attaining the status of a science. There were widespread false notions about how living organisms reproduce: bees were supposed to be born out of the carcass of an ox or a donkey (as described by the Roman poet Virgil); until Pasteur proved the contrary in the last century, many thought that flies and worms were engendered not by flies and worms but by rottenness; Aristotle had taught that the father provided the form of the child and the mother provided merely the matter; and so on.
The first quantitative study of inheritance appeared as a scientific paper in 1866, the author an Austrian monk, Gregor J. Mendel. But the results contained in it were not noticed or commented upon until about 1900. Mendel worked with pea-plants, cultivated in the monastery garden; these have the advantage that one can either pollinate a plant with the pollen of another or with its own. He listed seven characters or traits of these plants which are dichotomous, that is either-ors: (1) spherical seed Vs dented seed, (2) yellow seed vs. green seed, (3) purple flower vs. white flower, (4) inflated pod vs constricted pod, (5) green pod vs yellow pod, (6) axial flower vs terminal flower, and (7) tall stem vs dwarf stem. Then he bred fourteen populations or strains, each of which was "true-breeding" for each of these traits. This means that when, for example, two plants from the "true-breeding spherical seed" strain are crossed, the offspring has invariably spherical seeds. Similarly for the other traits: inside a given strain, fertilization always resulted in plants with the character or trait of that strain.
Next, Mendel took a plant from, say, the strain "yellow seed" and another from the strain "green seed," and pollinated one another (the male yellow and the female green and the other way around too). The parents are called the parental population P and the resulting offspring form the first filial population F1. Mendel observed that in F1 not a single plant had green seeds, they were all yellow!
He then proceeded to obtain a population F2 by pollinating the plants of F1 either with themselves or with each other. It turned out that about 1/4 of the plants in F2 had green seeds, the other 3/4 had yellow seeds. So, even though there were no green seeded plants in F1, the green ones reappeared in the second generation. To explain this, Mendel proposed the following model: the trait seed color, yellow or green, is associated with a pair of objects contained in the male and the female reproductory cells, which Mendel called "entities" and we call "genes." Yellow is associated with a gene Y and green with a gene g. The reason Y is capitalized and g is not is that Y and g are of a different kind: Mendel said Y is "dominant" and g is "recessive." Each plant, he reasoned, contains in its cells one of the following pairs: YY, Yg, gY, gg. Whenever the Y is present, the seeds will be yellow; so the only plants with green seeds are those who carry the genes gg. Of these pairs, a letter is contributed by each parent. This explains why the plants of the first generation F1 were all yellow seeded, since they had to carry the genes Y and g. Whereas for generation F2 we have the four equally likely possibilities, YY, Yg, gY and gg, and this is why roughly 1/4 of the plants in F2 were green-seeded. It's a question of probability, like tossing a coin, which gene will be contributed by each plant of F1, whether Y or g. It's important to remark that Mendel had never seen such genes; he postulated them to explain the observed ratios, and it turned out he was right (now we can see them with the electron microscope). Thus, what happened in genetics is very similar to what happened in physics and chemistry, where atoms and molecules were postulated (by Dalton and others) because they explained so well the phenomena, even though no one was able to perceive them.
Now, the same reasoning and the same ratios applied to the other six traits of pea plants: of each of them, one is dominant and the other recessive. For example, spherical seed is dominant and dented seed is recessive. Mendel also experimented with combined traits. Let us take, for example, the two pairs yellow seed (Y) vs green seed (g) and also spherical seed (S) vs dented seed (d). In the first generation, all the plants will have yellow, spherical seeds because they will all carry the genes Yg or gY and Sd or dS. But in the second generation, since there are four pairs YY, Yg, gY, gg, and also four pairs SS, Sd, dS, dd, we will have 4x4 = 16 combinations when we take into account both color and texture of the seed. Of these 16 combinations, 9 will have at least a Y and at least an S, so they will have yellow, spherical seeds; 3 will have no Y but at least an S, so the seeds will be green and spherical; 3 will have no S but at least a Y, so the seeds will be yellow and dented; and only 1 will have ggdd, yielding green, dented seeds. The expected proportions will then be 9:3:3:1. This is of course assuming that color and texture of the seeds are INDEPENDENT, as when we toss a coin twice or as when we toss two coins: the result of one toss doesn't affect the other in any way. And indeed, the numbers from Mendel's garden agreed (too well, as we'll see later) with the computed probabilities.
Two observations. First, this latter result with two traits combined, let us insist, will only be true if the traits are independent: today we know that means that the two genes are not in the same chromosome, since in reproduction parents transmit entire chromosomes rather than separate genes. And second, a great statistician, R. A. Fischer, showed about 1936 that Mendel's actual numbers were too close to the expected for comfort. Even when you toss a fair coin 10,000 times, it's extremely unlikely that you'll get exactly 5,000 heads. Well, Mendel's results, it turned out, were even more unlikely, and so, Fischer concluded, either Mendel fudged the data or he was extremely lucky. In any case, Mendel's theory is perfectly sound, and if he did fudge his data, he was neither the first nor the last eminent scientist to do so.
Let us give two examples of human traits whose transmission obeys rules similar (but not exactly equal) to those discovered by Mendel in pea plants. First, take hemophilia. This is a disease whose main symptom is that blood doesn't clot. It turns out that it is a sex-linked genetic trait. What this means is the following: any cell of a female contains two chromosomes (long chains of genes) called X, any cell of a male contains two chromosomes, one X and another Y. When a child is conceived, it inherits one of the Xs from the mother and either an X from the father (in which case the child is a girl) or a Y from the father (in which case the child is a boy). Now, the gene for hemophilia is located only in the X chromosome, and it is (fortunately!) recessive: if a man carries a hemophilic X chromosome, he suffers from hemophilia; a woman must carry two hemophilic X chromosomes to have the disease. Knowing this, we can figure the probabilities involved. Suppose a hemophilic man has a child with a non-carrier (normal) woman: then if the child is a girl she will be a carrier of hemophilia (because she has her father's X chromosome) but she will not manifest the disease herself; if the child is a boy he will be normal. Suppose now that a normal man has a child with a carrier woman: if the child is a girl, the chances are 1/2 and 1/2 that she'll be a carrier; if the child is a boy, the chances are 1/2 and 1/2 that he will actually suffer from hemophilia. The only way a girl can be hemophiliac is when her father is hemophiliac and her mother is a carrier, a highly unlikely event. Another example of a different kind is skin color in humans. It depends on several pairs of genes, and so it is called a "polygenetic" trait (of course, your skin color will not only depend on your genes but also on your exposure to the sun). A possible genetic model consists of three gene pairs Aa, Bb, and Cc, in which the capital letters do not mean dominant: they mean darkness of skin. A person will always have six of these genes, three inherited from the mother and three from the father. People with genotype AABBCC will have the darkest skin; people with genotype aabbcc will have the fairest; someone with AaBbCc will be just in the middle. Suppose we start with two "pure" populations, one with only AABBCC (darkest), and the other with aabbcc (fairest); then if we cross them, the first generation F1 will have people with genotype AaBbCc: they'll all have medium skin. The interesting thing is what happens with the people of F2, the second generation. For an individual in F2 the mother's contribution will depend on three choices: either A or a, either B or b and either C or c, so there are 2x2x2 = 8 choices. Similarly there are 8 choices for the father's contribution. So there will be 8x8 = 64 possible genotypes in F2. Of these, only one will be the fairest, aabbcc, and only one will be the darkest, AABBCC. But there are six genotypes containing just one capital letter (Aabbcc, aAbbcc, aaBbcc, aabBcc, aabbCc and aabbcC) and similarly there are also six containing five capital letters. There are 15 genotypes containing two capitals and also 15 containing four capitals, and there are 20 containing three capitals and three lower case (medium skin). The proportions are therefore: 1, 6, 15, 20, 15, 6, 1 (the total is 64). Notice that these are precisely the numbers appearing in Pascal's triangle, in row number 6. In particular, this means that if we graph the proportions of people with skin color ranging from fairest to darkest, the graph will resemble a bell curve!
These examples show the close relation between probability (the doctrine of chance) and heredity. This agrees with the saying that having children is a lottery: the child's traits will not depend on the will of the gods or the fairies, nor on the position of stars and planets at the time of the birth, but on something like the repeated flipping of a coin. We should remark that physics and biology have advanced along the same way: toward discrete models of reality, in which phenomena are explained on the basis of elementary units, and toward an ever wider application of probabilistic and statistical methods. But of modern physics we will talk in my next lecture.
Let us focus now on the physico-chemical mechanism of genes. As I said, Mendel had postulated them on the basis of his numerical data, but he had no idea of what they might look like. It was during this century, especially in the 50s and 60s, that the chemical structure of genetic material was discovered; new discoveries are being made even as we speak. To begin with, we are conceived—we burst into being—when two cells interact, an egg from our mother and a sperm from our father; by the time we reach adulthood there are about 1014 (a hundred trillion) cells in our body. From two to a hundred trillion: cells obviously reproduce, and as we will see, they do so in two different ways. The carrier of genetic information are large molecules of a substance called DNA, which are inside the nucleus of each cell. One or two such molecules form a chromosome; if there are two, they are united or pinched at a point called a centromere. Now, without going into detail, let us describe briefly how DNA carries genetic information and how cells reproduce.
DNA is the acronym for deoxyribonucleic acid; it is a long molecule formed by the union of shorter blocks of a similar (but not identical) kind: such long molecules are called "polymers" (Greek for "many parts"). Polymers are essential across life: proteins, carbohydrates like starch and cellulose and lipids (fatty substances) are all long chains with different building blocks. In starch and cellulose, for example, the building blocks are molecules of the sugar glucose, an asymmetric hexagonal molecule: the bonds between successive glucose rings are effected as an OH group hanging from one glucose attaches to another OH group hanging from another glucose, so that water (H2O) is liberated and a single O is left linking the two glucoses. In proteins, the building blocks are aminoacids, of which there are 20 different ones. With DNA (also with RNA, to be discussed below), the situation is as follows: the building blocks of DNA are the nucleic acids, which consist of three elements: (1) a sugar, deoxyribose, different from glucose in that it has a pentagonal molecule; (2) to this sugar there is attached a phosphate group (three phosphorus atoms plus oxygen), and (3) to another point of the sugar molecule there is attached a nitrogenous molecule called a base. Now the sugar and the phosphate group are always the same, and they provide the backbone of the polymer (the long molecule) since the sugar of one block becomes bonded to the phosphate of the next one and so on. But the nitrogen bases come in four different kinds: adenine (A), cytosine (C), guanine (G) and thymine (T). In 1953 Francis Crick and James Watson proposed a model for the DNA molecule: two strands or chains of nucleic acids forming a double helix. The two helices are attached to each other by the nitrogen bases in a very specific way: adenine pairs with thymine and guanine pairs with cytosine: A—T and C—G. Now the bonds between those nitrogen bases, and therefore between the two helices, are of a different kind from the bonds joining the backbones of the polymers: these backbones are joined by covalent bonds, where atoms share their electrons, and which is the way all small molecules are glued together; but the bonds between the bases of the two strands are "hydrogen bonds," the electrical attraction between molecules whose electrical charges are not symmetrically balanced, and this hydrogen bonds are about ten times weaker than covalent bonds. This is important because it makes possible for the two strands of a double helix to separate more easily.
Because of the rules for base bonding, A—T and C—G, knowing the sequence of bases in one strand of the double helix we automatically know the sequence in the other strand, by just replacing T for A and A for T, C for G and G for C. For DNA to reproduce itself, the two strands will essentially separate from each other and new strands will be built on the template of the given ones, thus creating two double helices where there was only one. Except for errors, the two new molecules of DNA will be identical to the original one. This is how DNA reproduces itself, but what's the use of DNA in the living organism? DNA is located in the nucleus of every cell of our body, and what it does is to direct the production of proteins in the cell, which amounts to saying that it governs all processes of cell life. The way DNA does that is rather complicated, and here we can offer only a brief summary. The information in DNA, that is the particular sequence of bases (for ex. AGCCGGTTATG... in humans the sequence is about 3 billion letters long!) is not used directly; rather it is used to build molecules of RNA, ribonucleic acid, which is another polymer. The main differences between DNA and RNA are the following: (1) RNA is only one strand rather than a double helix; (2) instead of the four bases in DNA (A, C, G, T), the last, thymine, is replaced by another base, uracil (U), so the rules for RNA base-pairing are: A—U, G—C; and (3) in the skeleton of RNA the sugar deoxyribose is replaced by ribose, another sugar. Incidentally, some viruses (like HIV) contain RNA but not DNA as genetic material. Anyway, in our own cells the information contained in DNA is used to build molecules of RNA, and these in turn are used to build molecules of proteins: a given long sequence of the four letters A, C, G, U, will cause the coming together of a certain sequence of amino acids, in other words, a long molecule of protein.
The replication of DNA is like the copying of a tape: we start with a double helix and end up with two. Based on this replication we have two mechanisms of cell division. The first is called mitosis. A cell will divide into two, and its DNA having replicated, each of these two daughter cells will have the same genetic material as the parent cell: if the latter had 2n chromosomes (our own cells have 46, that is, 23 pairs), the daughters will have 2n chromosomes each. In this way the daughter cells look exactly like the parent cell (clones). But there is another way of cell division called meiosis. Here, after DNA replication, the parent cell divides into four daughter cells, each of which has half the genetic material of the parent cell; so if this latter had 2n chromosomes, each of the four daughter cells will have n chromosomes. Meiosis is fundamental in sexual reproduction, for this is the way by which the sexual cells (gametes) are produced: each sexual cell contains only half the genetic material of the other cells. When two sexual cells, one from each parent, come together (fertilization), a new cell is created which has half of its DNA from each parent cell, and this genetic information is shuffled to create the DNA of the offspring.
These two fundamental ways of cell reproduction, mitosis and meiosis, correspond to two fundamental concepts of thought: sameness and difference. By mitosis we always get (except for chance errors) cells which are the same as the original one (clones). Meiosis, on the other hand, allows for chance shuffling of the genetic material of two cells to result in a new cell, different from either parent cell. Two observations are pertinent here. First, remember that last semester, talking about our own identity, I mentioned that it depends on memory. Without memory, I wouldn't know I am the same as a moment ago. But also, if my memory and my perception were perfect, I would not be able to say that I am the same person as a moment ago, since so much in my body, my perceptions and my mental contents has actually changed. So the consciousness of my identity and sameness require a memory that's less than perfect. Science too requires a perception and a memory that are less than perfect, otherwise two cells or two molecules would never be the same, and no scientific laws would be possible. Then what about difference? Science, which is based on causality, has always had a problem with explaining what's different, what's new. For if any future event can be predicted from present and past circumstances by the application of exact mathematical laws, everything would be predetermined and nothing would be genuinely new. The way science deals with this problem is by using the concept of chance. Chance events cannot be exactly predicted, only approximate probabilities can be computed. And as we saw, biology explains the birth of the new, the unique human being that each of us is, by placing chance at the origin of life: we are the product of a random shuffling of DNA at the moment of conception. Professor Isser talked about the recent cloning of animals, and this may be viewed as still another attempt by humans to control the working of chance: cloning means reproduction by mitosis, so that chancy sexual reproduction is set aside.
Prof. Isser also mentioned in his lectures two philosophers, Nietzsche and Heidegger. Here's what Nietzsche had to say, in the 1880s, about scientific and technological advances (such as cloning): "We are attempting a venture with truth! Perhaps humanity will perish by it! So be it!" And here's what Heidegger had to say, in 1954: "Since man is the most important raw material, one can reckon with the fact that some day factories will be built for the artificial breeding of human material, based on present-day chemical research. ... The way in which artificial insemination is handled corresponds with stark consistency to the way in which literature is handled in the sector of `culture'. (Let us not flee because of antiquated prudery to distinctions which no longer exist. The need for human material underlies the same regulation of preparing for ordered mobilization as the need for entertaining books and poems, for whose production the poet is no more important than the bookbinder's apprentice, who helps bind the poems for the printer by, for example, bringing the covers for binding from the storage room.)" Depressing? Yes. But I didn't want to end this lecture about the marvelous advances in scientific research without telling you about some of the frightened and frightening reactions it has elicited.