Chapter 13: Chromosomes PDF

CHAPTER 13: CHROMOSOMES 13.1 Portrait of a Chromosome Mutations range from single-base changes to entire extra sets of chromosomes. A mutation is considered a chromosomal aberration if it is large enough to be seen with a light microscope using stains and/or fluorescent probes to highlight...

CHAPTER 13: CHROMOSOMES 13.1 Portrait of a Chromosome Mutations range from single-base changes to entire extra sets of chromosomes. A mutation is considered a chromosomal aberration if it is large enough to be seen with a light microscope using stains and/or fluorescent probes to highlight miss-ing,extra,or moved DNA sequences. A chromosome includes hundreds to thousands of genes. In general,too little genetic material has more severe effects on health than too much. Extensive chromosome abnor-malities that are present in all cells of an embryo or fetus dis-rupt or halt prenatal development. As a result, only 0.65 percent of all newborns have chromosomal abnormalities that producesymptoms.An additional 0.20 percent of newborns have chro-mosomal rearrangements in which chromosome parts have flipped or swapped,but the rearrangements do not produce symptoms unless they disrupt the structures or functions of genes that affect health. Cytogenetics is the classical area of genetics that links chromosome variations to specific traits, including illnesses.This chapter explores several ways that chromosomes can be atypical (used synonymously with abnormal) and affect health.Actual cases introduce some of them. Required Parts:Telomeres and Centromeres A chromosome consists primarily of DNA and proteins with a small amount of RNA. Chromosomes are duplicated and transmitted,via cell division (mitosis or meiosis),to the nextcell generation. The different chromosome types have long been described and distinguished by size and shape,using stains and dyes to contrast dark heterochromatin with the lighter euchromatin (figure 13.1).Heterochromatin consists mostly of highly repetitive DNA sequences,whereas euchro-matin has many protein-encoding sequences. A chromosome must include structures that enable it to replicate and remain intact. Everything else is informational: protein-encoding genes and their controls.The essential parts of a chromosome are: telomeres; origin of replication sites, where replication forks begin to form;and the centromere. Recall from figure 2.15 that telomeres are chromosome tips. In humans,each telomere repeats the sequence TTAGGG.In most cell types, telomeres shorten with each mitotic cell division. The centromere is the largest constriction of a chromo-some.It is where spindle fibers attach when the cell divides.A chromosome without a centromere is no longer a chromo-some. It vanishes from the cell as soon as division begins because there is no way to attach to the spindle. Centromeres,like chromosomes,consist mostly of DNAand protein. Many of the hundreds of thousands of DNA bases that form the centromere are copies of a specific 171-baseDNA sequence.The size and number of repeats are similar in many species,although the sequence differs.The similar-ity among species suggests that these repeats have a structural role in maintaining chromosomes rather than an informational one from their sequence.Certain centromere-associated proteins are synthesized only when mitosis is imminent,forming a structure called a kinetochore that contacts the spindle fibers,enabling the cell to divide. Centromeres replicate toward the end of S phase of the cell cycle.A protein that may control the process is centromere protein A, or CENP-A. Molecules of CENP-A stay with centromeres as chromosomes replicate, covering about half a million DNA base pairs. When the replicated (sister) chromatids separate at anaphase, each member of the pair retains some CENP-A.  The protein therefore passes to the next cell generation, but it is not DNA. This is an epigenetic change. Centromeres lie within vast stretches of heterochromatin. The arms of the chromosome extend outward from the centro-mere. Gradually, with increasing distance from the centromere, the DNA includes more protein-encoding sequences. Gene density varies greatly among chromosomes. Chromosome 21 includes a gene "desert," harboring a million-base stretch with no protein-encoding genes at all. Chromosome 22, in con-trast, is a gene "jungle." Yet these two tiniest chromosomes are remarkably similar in size, which is why they are out of place in karyotypes. Chromosome 22 contains 545 genes to chromosome 21's 225! That is, chromosome 21 is actually the smallest, if size is measured by gene content. The chromosome parts that lie between protein-rich areas and the telomeres are termed subtelomeres (figure 13.2). These areas extend from 8,000 to 300,000 bases inward toward the centromere from the telomeres. Subtelo-meres include some protein-encoding genes and therefore bridge the gene-rich regions and the telomere repeats. The transition is gradual. Areas of 50 to 250 bases, right next to the telomeres, consist of 6-base repeats, many of them very similar to the TAGGG of the telomeres. Then, moving inward from the 6-base zone are many shorter repeats. Their function isn't known. Finally, the sequence diversifies and protein-encoding genes appear. At least 500 protein-encoding genes lie in the subtelo-mere regions of all the chromosomes combined. About half are members of multigene families (groups of genes of very similar sequence next to each other) that include pseudogenes. These multigene families may be signs of recent evolution: Apes and chimps have only one or two genes for many of the large gene families in humans. Karyotypes Chart Chromosomes An old but still useful tool is the chromosome chart, or karyotype. By showing which genes are transmitted together as part of the same chromosome, a karyotype can reveal certain conditions that DNA sequencing can miss. A karyotype displays chromosomes in pairs by size and by physical landmarks that appear during mitotic metaphase, when DNA coils tightly, enabling it to be visualized. Figure 13.3 shows a karyotype with one extra chromosome, which is called a trisomy. The 24 human chromosome types are numbered from largest to smallest—1 to 22. The other two chromosomes are the X and the Y. Early attempts to size-order chromosomes resulted in generalized groupings because many of the chromosomes are of similar size. Use of dyes and stains made it easier to distinguish chromosomes by creating patterns of bands. Centromere position is one physical feature of chro-mosomes. A chromosome is metacentric if the centromere divides it into two arms of approximately equal length. It is submetacentric if the centromere establishes one long arm and one short arm, and acrocentric if it pinches off only a small amount of material toward one end (figure 13.4). Some species have telocentric chromosomes that have only one arm, but humans do not. The long arm of a chromosome is designated g, and the short arm p (for "petite"). Five human chromosomes (13, 14, 15, 21, and 22) have bloblike ends, called satellites, that extend from a thin, stalklike bridge. The stalk regions do not stain, but they carry many copies of genes encoding ribosomal RNA and ribosomal proteins. These areas coalesce to form the nucleolus, a structure in the nucleus where ribosomal building blocks are produced and assembled (see figure 2.2). Karyotypes are useful at several levels. When a baby is born with distinctive facial characteristics of a chromosomal syndrome, a karyotype can confirm the clinical diagnosis. Within families, karyotypes are used to identify relatives with a chromosome aberration that can affect health. In one family, several adults died from a rare form of kidney cancer. Karyotypes revealed that the affected individuals all had an exchange of genetic material between chromosomes 3 and 8. When karyotypes showed that two healthy young family members had the unusual chromosomes, further testing indicated the cancer. They were successfully treated. Karyotypes of individuals from different populations can reveal the effects of environmental toxins, if abnormalities appear only in a group exposed to a contaminant. Because chemicals and radiation that can cause cancer and birth defects often break chromosomes into fragments or rings, detecting this genetic damage can alert researchers to the possibility that certain cancers may appear in the population. 13.2 Detecting Chromosomes Chromosomes can be imaged from any type of human cell that has a nucleus (red blood cells do not). The most common application of human chromosome testing is in prenatal diagnosis. Chromosomes are also checked in relatives of people known to have atypical chromosomes, to explain infertility, or to diagnose or track changes as a cancer progresses. Older Techniques The technologies that identify atypical chromosomes in fetuses have evolved since the first fetal karyotype was constructed in 1966. Just in the past few years, deducing chromosome counts from short pieces of placental DNA floating in the maternal bloodstream has mostly replaced the older approaches of amniocentesis and chorionic villus sampling (CVS), which construct karyotypes from cells from structures that support the fetus. In amniocentesis, a needle passed through the abdominal wall removes a small sample of amniotic fluid from the uterus of a pregnant woman (figure 13.5a). Fetal cells in the fluid are cultured for 7 to 10 days, and then 20 cells are karyotyped. Additional tests on biochemicals in the fluid and single-gene tests can reveal some inborn errors of metabo-lism. Ultrasound is used to place the needle and to visualize the fetus (figure 13.5c). Amniocentesis can be safely performed after 14 weeks. For many years amniocentesis was limited to women over age 35, when the risk of the procedure equals the risk of miscarriage, which rises with maternal age (figure 13.6). As safety improved, amniocentesis was offered to younger women. Then other tests were developed. The cells sampled between weeks 10 and 12 in CVS come from the chorionic villi, which are fingerlike structures that develop into the placenta (figure 13.5b). Because chorionic villi cells and fetal cells descend from the fertilized ovum, it is assumed that they will have the same chromosomal content. However, rarely, a mutation can occur in either type of cell, leading to a false negative or false positive. If CVS indicates an abnormal chromosome in a villus cell that is not also in the fetus, the couple may elect termination of a chromosomally normal fetus. In the opposite situation, a chorionic villus cell might have normal chromosomes when the fetus does not, providing a false sense of security that all is well. CVS cannot detect inborn errors in biochemicals because amniotic fluid is not sampled. As the techniques for collecting chromosomes evolved, so did the methods for distinguishing them. Early karyotypes used generalized stains, which could not distinguish chromosomes of similar size, which were considered in groups. Development of more discriminating binding chemicals led to the ability to tell chromosomes apart by banding patterns. A technique called fluorescence in situ hybridization (FISH) adds precision by applying pieces of DNA, called probes, which are attached to molecules that produce a flash of color when they bind their complements on the chromosomes in a tissue sample. Use of an algorithm applies a unique false color to each chromosome, "painting" a karyotype from several fluorescent dyes. Many laboratories use FISH probes for the most common chromosomal abnormalities: an extra chromosome 13, 18, or 21, and sex chromosome anomalies. Figure 13.7 shows the extra flash of a cell from a person with trisomy 21. Cell-Free Fetal DNA Testing Small pieces of DNA are normally present in the blood-stream. In a pregnant woman, about 20 percent of those pieces come from the placenta, and therefore represent the fetus (figure 13.8). The DNA pieces are called "cell-free" because they are not enclosed in cells. Tests of cell-free fetal DNA are done at 10 weeks or later, and are rapidly replacing the older techniques. It is also called noninvasive prenatal diagnosis or testing. (Ultrasound is also noninvasive.) Analysis of cell-free fetal DNA is based on proportions. For example, blood from a woman whose fetus has trisomy 21 Down syndrome has about 50 percent more DNA pieces from chromosome 21 than it does from the other chromosomes.That extra DNA represents the third copy of the chromosome. Amniocentesis may be offered to confirm a finding using cell-free DNA, from which a karyotype is constructed showing the extra chromosome. One advantage of the cell-free analysis is that women of all ages use it, saving older women from the more invasive older techniques, while also detecting chromosomal conditions in the fetuses of younger women that might have been missed with the older techniques. Cell-free testing is so sensitive that it has even picked up cancer cells, which also shed DNA, in women who do not yet have cancer symptoms. To detect single-gene conditions, cell-free fetal DNA must be compared to haplotypes from blood samples from the parents for the same chromosomal regions. This is done to distinguish mutations that are in the fetus from those in the pregnant woman. For example, a fetus with identical mutant CFTR alleles yet no wild type alleles may have indeed inherited cystic fibrosis from two carrier parents, but the sample might only have captured mutant alleles from the pregnant woman. A haplotype would show surrounding DNA sequences, enabling the distinction between the fetal and maternal genomes. Some laboratories analyze cell-free fetal DNA only for the more common chromosomal abnormalities, but entire genomes can be reconstructed from the DNA pieces. These more complete analyses are used to distinguish de novo (new) mutations from inherited mutations in parent-child trios, such as the family described in Clinical Connection 1.1. An indirect type of test indicates an elevated risk of having an abnormal chromosome number. It measures biochemi-cals whose levels in the blood are within a certain range in a pregnant woman carrying a fetus with the normal number of chromosomes, but lie outside that range in fetuses whose cells have an extra copy of a certain chromosome. Chromosomal Shorthand The information in a karyotype is abbreviated by listing the chromosome number, sex chromosome makeup, and atypical autosomes (discussed in subsequent sections of this chapter) (table 13.1). Symbols and shorthand describe the type of aber-ration, such as "del" for deletion and "I" for translocation, in which chromosomes exchange parts or a piece of one chromosome is moved to another. After the shorthand, numbers correspond to bands and subbands that are assigned to the markings on chromosomes that arise from staining and FISH. A chromosomally normal male is designated 46,XY and a female 46,XX. Bands and subbands identify specific genes, For example, the gene that encodes the beta-globin subunit of hemoglobin is located at 11p15.5, the short arm of chromosome 11 at subband 15.5. 13.2 Atypical Chromosome Number  A human karyotype is atypical (abnormal) if the number of chromosomes in a somatic cell is not 46, or if individual chromosomes have extra, missing, or rearranged genetic material. More discriminating technologies can detect very small numbers of extra or missing nucleotides. As a result, more people are being diagnosed with chromosomal abnormalities than in the days when stains made all chromosomes look alike. Atypical chromosomes account for at least 50 percent of spontaneous abortions, yet only 0.65 percent of newborns have them. Therefore, most embryos and fetuses with atypical chromosomes stop developing before birth. Table 13.2 summarizes the types of chromosome variants in the order in which they are discussed. Polyploidy The most extreme upset in chromosome number is an entire extra set. A cell with extra sets of chromosomes is polyploid. An individual whose cells have three copies of each chromosome is a triploid (designated 3N, for three sets of chromo-somes). Two-thirds of all triploids result from fertilization of an oocyte by two sperm. The other cases arise from formation of a diploid gamete, such as when a normal haploid sperm fertilizes a diploid oocyte. Triploids account for 17 percent of spontaneous abortions (figure 13.10). Very rarely, an infant survives a few days, with defects in nearly all organs. However, certain human cells may be polyploid. The liver, for example, has some tetraploid (4N) and even octaploid (8N) cells. Polyploids, although uncommon in humans, are common among flowering plants, including roses, cotton, barley, and wheat, and in some insects. Fish farmers raise triploid salmon, which cannot breed because their gametes contain different numbers of chromosomes. Aneuploidy Cells missing a single chromosome or having an extra chromosome are aneuploid, which means "not good set." Rarely, aneuploids can have more than one missing or extra chromo-some, indicating abnormal meiosis in a parent. A normal chromosome number is euploid, which means "good set." Most autosomal aneuploids (with a missing or extra non-sex chromosome) are spontaneously aborted. Those that survive have specific syndromes, with symptoms depending on which chromosomes are missing or extra. Intellectual disability is common in aneuploidy because development of the brain is so complex and of such long duration that nearly any chromosome-scale disruption affects genes whose protein products affect the brain. Sex chromosome aneuploidy usually produces milder symptoms. Most children born with a chromosome number other than 46 have an extra chromosome (a trisomy) rather than a missing one (a monosomy), because monosomies are typically so severe that an affected embryo ceases developing (except some cases of a female with only one X chromosome). Trisomies and monosomies are named for the chromosomes involved, and in the past the associated syndromes were named for the discoverers. Today, cytogenetic terminology is used because it is more precise. For example, Down syndrome can result from a trisomy or a translocation. The distinction is important in genetic counseling. Translocation Down syndrome, although accounting for only 4 percent of cases, has a much higher recurrence risk within a family than does trisomy 21 Down syndrome, a point we return to later in the chapter. The meiotic error that causes aneuploidy is called nondisjunction. Recall that in normal meiosis, homologs separate and each of the resulting gametes receives only one member of each chromosome pair. In nondisjunction, a chromosome pair does not separate at anaphase of either the first or second meiotic division. This unequal division produces a sperm or oocyte that has two copies of a particular chromo-some, or none, rather than one copy (figure 13.11). When such a gamete meets its partner at fertilization, the resulting zygote has either 45 or 47 chromosomes, instead of the normal 46. Different trisomies tend to be caused by nondisjunction in the male or female, at meiosis I or II. A cell can have a missing or extra chromosome in 49 ways—an extra or missing copy of each of the 22 autosomes, plus the sex chromosome combinations of Y, X, XXX, XXY, and XYY. (Some individuals have four or even five sex chromo-somes.) However, only nine types of aneuploids are recognized in newborns. Others are seen in spontaneous abortions or fertilized ova intended for in vitro fertilization. Most of the 50 percent of spontaneous abortions that result from extra or missing chromosomes are 45,X individuals (miss-ing an X chromosome), triploids, or trisomy 16. About 9 percent of spontaneous abortions are trisomy 13, 18, or 21. More than 95 percent of newborns with atypical chromosome numbers have an extra chromosome 13, 18, or 21, or an extra or missing X or Y chromosome. These conditions are all rare at birth— together they affect only 0.1 percent of all children; however, nondis-junction occurs in 5 percent of recognized pregnancies. Types of chromosome abnormalities differ between the sexes. Atypical oocytes mostly have extra or missing chromo-somes, whereas atypical sperm more often have structural vari-ants, such as inversions or translocations, discussed later in the chapter. Aneuploidy and polyploidy also arise during mitosis, producing groups of somatic cells with the extra or missing chromosome. A mitotic abnormality that occurs early in development, so that many cells descend from the unusual one, can affect health. For example, a chromosomal mosaic for a trisomy may have a mild version of the associated condition. This is usually the case for 1 to 2 percent of people with Down syndrome; they are mosaics. The phenotype depends on which cells have the extra chromosome. Unfortunately, prenatal testing cannot reveal which cells are affected. Autosomal Aneuploids Most autosomal aneuploids cease developing long before birth. Following are cases and descriptions of the most common autosomal aneuploids among liveborns. The most frequently seen extra autosomes in newborns are chromosomes 21, 18, and 13, because these chromosomes carry many fewer. protein-encoding genes than the other autosomes, compared to their total amount of DNA. Therefore, extra copies of these chromosomes are tolerated well enough for some fetuses with them to survive to be born (table 13.3).. The most common autosomal aneuploid among liveborns is trisomy 21, because this chromosome has the fewest genes. On an ultrasound scan, a fetus with Down syndrome has short limbs, a flattened nose, and excess fluid at the back of the neck. A person with Down syndrome is usually short and has straight, sparse hair and a tongue protruding through thick lips. The hands have an atypical pattern of creases, the joints are loose, and poor reflexes and muscle tone give a "floppy" appearance to a baby. Developmental milestones (such as sitting, standing, and walking) come slowly, and toilet training may take several years. Intelligence varies greatly. Parents of a child with Down syndrome can help their child reach maximal potential by providing a stimulating environment (figure 13.12). Many people with Down syndrome have physical prob-lems, including heart and kidney defects and hearing and visual loss. A suppressed immune system can make influenza deadly.cDigestive system blockages are common and may require surgical correction. A child with Down syndrome is 15 times more likely to develop leukemia than a child who does not have the syndrome, but this is still only a 1 percent risk. However, people with Down syndrome are somewhat protected against developing solid tumors. Many of the medical problems associated with Down syndrome are treatable, so that average life expectancy is now 60 years. In 1910, life expectancy was only 9 years. Down syndrome occurs in all population groups. Some people with Down syndrome older than 40 develop the black tau fibers and amyloid beta protein deposits in their brains characteristic of Alzheimer disease, although they usually do not develop severe dementia (see Clinical Connection 5.1). The chance of a person with trisomy 21 developing Alzheimer disease is 25 percent, compared to 6 percent for the general pop-ulation. A gene on chromosome 21 causes one inherited form of Alzheimer disease. Perhaps the extra copy of the gene in trisomy 21 has a similar effect to a mutation in the gene that causes Alzheimer disease, such as causing amyloid beta buildup. Before human genome sequencing began, researchers studied people who have a third copy of only part of chromosome 21 to deduce which genes cause which symptoms. This approach led to the discovery that genes near the tip of the long arm of the chromosome contribute most of the abnormalities. Two specific genes control many aspects of Down syndrome by controlling a third gene, which encodes a transcription factor and therefore affects expression of many other genes. A newer approach, genome editing, is being used to study Down syndrome at the cellular level. Section 21.4 discusses how genome editing cuts double-stranded DNA at a specific site on a chromosome and allows insertion of a specific DNA sequence. For trisomy 21 Down syndrome, the technique "bor-rows" the mechanism that shuts off one X chromosome in the cells of females. Researchers insert the DNA that encodes the long noncoding RNA sequence called XIST, which normally shuts off one X chromosome (see figure 6.9), into one chromosome 21 of induced pluripotent stem cells (see Table 2.3) made from skin cells of a boy with trisomy 21. The shut-off chromosome 21 forms a Barr body, which normally happens when XIST silences one X chromosome in a female cell. (Male cells do not normally have Barr bodies because they have only - one X chromosome.) The treated male cells have Barr bodies (figure 13.13), meaning one of the three copies of chromosome 21 is turned off. This experiment at the cellular level may lead to new treatments. The likelihood of giving birth to a child with trisomy 21 Down syndrome increases dramatically with the age of the mother (see figure 13.6). However, 80 percent of children with trisomy 21 are born to women under age 35. About 90 percent of trisomy 21 conceptions are due to nondisjunction during meiosis I in the female. The 10 percent of cases due to the male result from nondisjunction during meiosis I or II. The chance that trisomy 21 will recur in a family, based on empirical data (how often it actually does recur in families, is 1 percent. The age factor in trisomy 21 Down syndrome and other trisomies may be because the older a woman is, the longer her oocytes have been arrested on the brink of completing meiosis. During this 15 to 45 years, oocytes may have been exposed to toxins, viruses, and radiation. A second explanation for the maternal age effect is that females have a pool of immature aneuploid oocytes resulting from spindle abnormalities that cause nondisjunction. As a woman ages, selectively releasing normal oocytes each month, the atypical ones remain. The association between maternal age and Down syndrome has been recognized for a long time, because affected individuals were often the youngest children in large fami-lies. Before the chromosome connection was made in 1959, the syndrome was attributed to syphilis, tuberculosis, thyroid malfunction, alcoholism, emotional trauma, or "maternal reproductive exhaustion." The increased risk of Down syndrome correlates to maternal age, not to the number of children in the family. Bioethics examines the effect of prenatal diagnosis on the prevalence of people with trisomy 21 Down syndrome. Trisomy 18 Most individuals with trisomy 18 (Edwards syndrome) or trisomy 13 (Patau syndrome) are not born or die in infancy, but a few have lived into young adulthood. Most children who have trisomy 18 have great physical and intellectual disabilities, with developmental skills stalled at the 6-month level. Major abnormalities include heart defects, a displaced liver, growth retardation, and oddly clenched fists. Milder signs include overlapping fingers, a narrow and flat skull, low-set ears, a small mouth, unusual fingerprints, and "rocker-bottom" feet. Most cases of trisomy 18 arise from nondisjunction in meiosis Il of the oocyte. Trisomy 13 Trisomy 13 has different signs and symptoms than trisomy 18. Most striking is fusion of the developing eyes into one large eyelike structure in the center of the face, or a small or absent eye. Major abnormalities affect the heart, kidneys, brain, face, and limbs. The nose is often malformed, and cleft lip and/or palate is present in a small head. There may be extra fingers and toes. Ultrasound examination of an affected newborn may reveal an extra spleen, atypical liver, rotated intestines, and an abnormal pancreas. Sex Chromosome Aneuploids: Female People with sex chromosome aneuploidy have extra or missing sex chromosomes. Table 13.4 indicates how these aneuploids can arise. Some conditions can result from nondisjunction in meiosis in the male or female. Sex chromosome aneuploids are generally associated with much less severe symptoms and characteristics than autosomal aneuploids. The syndromes are no longer commonly referred to with eponyms, but 47,XXY was called Klinefelter syndrome, 45,X was Turner syndrome (also called 45,XO), and 47,XYY was Jacobs syndrome. 45,X Syndrome Like the autosomal aneuploids, 45, syndrome is more frequent among spontaneously aborted fetuses than among newborns—99 percent of 45,X fetuses are not born. The syndrome affects 1 in 2,500 female births. However, if amniocentesis or CVS was not done, a person with 45,X syndrome would likely not know she has a chromosome abnormality until she lags in sexual development and has her chromosomes checked. Two X chromosomes are necessary for normal sexual development in females. At birth, a girl with 45,X syndrome looks normal, except for puffy hands and feet caused by impaired lymph flow. In childhood, signs of 45,X syndrome include wide-set nipples, soft nails that turn up at the tips, slight webbing at the back of the neck, short stature, coarse facial features, and a low hairline at the back of the head. About half of people with 45X syndrome have impaired hearing and frequent ear infections due to a small defect in the shape of the coiled part of the inner ear. They cannot hear certain frequencies of sound. At sexual maturity, sparse body hair develops, but the girls do not ovulate or menstruate, and their breasts do not develop. The uterus is very small, but the vagina and cervix are normal size. In the ovaries, oocytes mature too fast, depleting the supply during infancy. 45, syndrome may impair the ability to solve math problems that entail envisioning objects in three-dimensional space, and may cause memory deficits, but intelligence is normal. Low doses of hormones (estrogen and progesterone) can stimulate development of secondary sexual structures for girls diagnosed before puberty, and growth hormone can maximize height. Individuals who are mosaics (only some cells are 45,X) may have children, but their offspring are at high risk of having too many or too few chromosomes. 45,X syndrome is unrelated to the age of the mother. The effects of 45,X syndrome continue past the reproductive years. Adults with the syndrome are more likely to develop certain diseases than the general population, including osteoporosis, types 1 and 2 diabetes, and colon can-cer. The many signs and symptoms of 45,X syndrome result from loss of specific genes. Life span is shortened slightly. Triplo-X About 1 in every 1,000 females has an extra X chromosome in each of her cells, a condition called triplo-X. The only symptoms are tall stature and menstrual irregularities. Although triplo-X females are rarely intellectually disabled, they tend to be less intelligent than their siblings who have the normal number of chromosomes. The lack of symptoms reflects the protective effect of X inactivation-all but one of the X chromosomes is inactivated. Sex Chromosome Aneuploids: Male Any individual with a Y chromosome is a male. A man can have one or more extra X or Y chromosomes. 47,XX7 Syndrome About 1 in 500 males has the extra X chromosome that causes 47,XXY syndrome. Severely affected men are underdeveloped sexually, with rudimentary testes and prostate glands and sparse pubic and facial hair. They have very long arms and legs, large hands and feet, and may develop breast tissue. Yet some men with an extra X chromosome have no associated symptoms. Testosterone injections during adolescence can limit limb lengthening and stimulate development of secondary sexual characteristics. Boys and men with 47,XXY syndrome may be slow to learn, but they are usually not intellectually disabled unless they have more than two X chromosomes, which is rare. 47,XXY syndrome is the most common genetic or chromosomal cause of male infertility. Doctors can select sperm that contain only one sex chromosome and use the sperm to fertilize oocytes in vitro. However, sperm from men with 47,XXY syndrome are more likely to have extra chromosomes—usually X or Y, but also autosomes—than sperm from men who do not have 47,XXY syndrome. 47,XXYY Syndrome About 1 in 17,000 newborn boys have an extra X chromosome and an extra Y chromosome. These 48,XXYY males have more severe behavioral problems than males with 47,XXY syndrome and tend to develop foot and leg ulcers, resulting from poor venous circulation. Attention deficit disorder, obsessive compulsive disorder, autism, and learning disabilities typically develop by adolescence. In the teen years, testosterone level is low, development of secondary sexual characteristics is delayed, and the testes are undescended. A man with 48,XXYY syndrome is infertile. 47,XYY Syndrome One male in 1,000 has an extra Y chromosome. Today, we know that 96 percent of males with an extra Y chromosome are not destined to become criminals, but may be tall and have acne. Problems with speech and understanding language are subtle, such as an inability to understand humor. Attaining skills like dressing and socializing may be delayed. The higher prevalence of 47,XYY among mental and penal institution populations may stem more from psychology than biology. Large body size may lead teachers, employers, parents, and others to expect more of these people, and a few 47,XYY individuals may deal with this stress aggressively. 47,XYY syndrome arises from nondisjunction in the male, producing a sperm with two Y chromosomes that fertilizes a normal oocyte. Geneticists have never observed a sex chromosome constitution of one Y and no X because the X chromosome carries so many genes that a 45,X embryo stops developing very early. 13.4 Atypical Chromosome Structure  A chromosome can be structurally atypical in several ways. It may have too much or too little genetic material, or a stretch of DNA that is inverted or moved and inserted into a different type of chromosome (figure 13.15). Atypical chromosomes are balanced if they have the normal amount of genetic material and unbalanced if they have extra or missing DNA sequences. Deletions and Duplications Deletion and duplication mutations are missing and extra DNA sequences, respectively. They are types of copy number variants (CNVs). (Sections 7.3 and 12.5 discussed CNVs.) The more genes involved, the more severe the associated syndrome. Figure 13.16 depicts a common duplication, of part of chromosome 15. Many deletions and duplications are de novo (neither parent has the abnormality). A technique called comparative genomic hybridization (CGH) is used to detect very small CNVs, which are also termed microdeletions and microduplications. CGH compares the number of copies of a CNV in the same amount of DNA from two people—one with a medical condition, one healthy, to associate the duplication or deletion with the phenotype. CGH is used to help diagnose autism, intellectual disability, learning disabilities, and other behavioral conditions. For example, the technique showed that a young boy who had difficulty concentrating and sleeping and would often scream for no apparent reason had a small duplication in chromosome 7. A young girl plagued with head-banging behavior, digestive difficulties, severe constipation, and great sensitivity to sound had a microdeletion in chromosome 16. Other microdeletions cause male infertility. Deletions and duplications can arise from chromosome rearrangements. These include translocations, inversions, and ring chromosomes. Translocation Down Syndrome In a translocation, different (nonhomologous) chromosomes exchange or combine parts. Translocations can be inherited because they can be present in carriers, who have the normal amount of genetic material, but it is rearranged. A translocation can affect the phenotype if it breaks a gene or leads to duplications or deletions in the chromosomes of offspring. There are two major types of translocations, as well as a few rarer types. In a Robertsonian translocation, the short arms of two different acrocentric chromosomes break, leaving sticky ends on the two long arms that join, forming a single, large chromosome with two long arms (see chromosome 14/21 in figure 13.17). The tiny short arms are lost, but their DNA sequences are repeated elsewhere in the genome, so the loss does not cause symptoms. The person with the large, translocated chromosome, called a translocation carrier, has 45 chromosomes instead of 46, but may not have symptoms if no crucial genes have been deleted or damaged. Even so, he or she may produce unbalanced gametes—sperm or oocytes with too many or too few genes. This can lead to spontaneous abortion or birth defects. In 1 out of 20 cases of Down syndrome, a parent has a Robertsonian translocation between chromosome 21 and another, usually chromosome 14. That parent produces some gametes that do not have either of the involved chromosomes and some gametes that have extra material from one of the translocated chromosomes. In such a case, each fertilized ovum has a 1 in 2 chance of being spontaneously aborted, and a 1 in 6 chance of developing into an individual with Down syndrome. The risk of giving birth to a child with Down syndrome is theoretically 1 in 3, because the spontaneous abortions are not births. However, because some Down syndrome fetuses spontaneously abort, the empiric risk of a couple in this situation having a child with Down syndrome is about 15 percent. The other two outcomes—a fetus with normal chromosomes or a translocation carrier like the parent—have normal phenotypes. Either a male or a female can be a translocation carrier, and the condition is not related to age. About 1 in 1,000 individuals in the general population carries a Robertsonian translocation. Much rarer than being a heterozygote for a Robertsonian translocation is being a homozygote, because such individuals can only arise from inheriting one copy of the unusual chromosome from each parent—which typically means the parents are related and inherited the translocation from a shared ancestor, like a common great-grandparent. Robertsonian homozygotes have 44 chromosomes rather than the normal 46. A case report describes a healthy 25-year-old man from China who has 44 chromosomes because each chromosome 14 joins a chromosome 15. His parents, both translocation carriers, were first cousins. The Chinese man’s sperm carry 21 autosomes and an X or Y. He can only be fertile with a woman whose cells have the same 44 chromosome types. Robertsonian translocation homozygotes bring up an interesting scenario. If several people with the same 44 chromosome types breed among themselves, they could, theoretically, start a new human subspecies. The probability of two carriers of the same Robertsonian translocation randomly meeting and having children together is about 1 in 4 million. Clinical Connection 13.1 describes how genomewide SNP analysis of sequence variants can miss a Robertsonian translocation if all of the parts of the affected chromosomes are present. In a reciprocal translocation, the second major type of translocation, two different chromosomes exchange parts (figure 13.18). About 1 in 600 people is a carrier for a reciprocal translocation. FISH can be used to highlight the involved chromosomes. If the chromosome exchange does not break any genes, then a person who has both translocated chromosomes is healthy and a translocation carrier. He or she has the normal amount of genetic material, but it is rearranged. A reciprocal translocation carrier can have symptoms if one of the two breakpoints lies in a gene, disrupting its function. For example, a translocation between chromosomes 11 and 22 causes infertility in males and recurrent pregnancy loss in females. However, a reciprocal translocation that arises de novo in a sperm or oocyte can affect health in the next generation if fertilization occurs and development proceeds, because the resulting individual would have extra or missing genetic material. Reciprocal translocations usually occur in specific chromosomes that have unstable parts. Vulnerable parts of chromosomes are located where the DNA is so symmetrical in sequence that complementary base pairing occurs within the same DNA strand, folding it into loops during DNA replication. When these contortions break both DNA strands, parts of different chromosomes can switch places. Unbalanced gametes can result when a reciprocal translocation is inherited or de novo, just as they can from a Robertsonian translocation. The four possibilities are (1) transmitting two normal copies of the two involved chromosomes; (2) transmitting two abnormal copies, with no effect on the phenotype of an offspring unless a vital gene is disrupted; or (3 and 4) transmitting either translocated chromosome, which introduces extra or missing genetic material and likely would affect the phenotype. A rare type of translocation is an insertional translocation, in which part of one chromosome inserts into a nonhomologous chromosome. Symptoms may result if the inserted DNA disrupts a vital gene or if crucial DNA sequences are lost or present in excess. A carrier of any type of translocation can produce unbalanced gametes—sperm or oocytes that have deletions or duplications of some of the genes in the translocated chromosomes. The phenotype depends on the genes that the rearrangement disrupts and whether genes are extra or missing. A translocation and a deletion can cause the same syndrome if they affect the same part of a chromosome. A genetic counselor suspects a translocation when a family has a history of birth defects, pregnancy loss, and/or stillbirths. Inversions An inverted sequence of chromosome bands disrupts important genes and harms health in only 5 to 10 percent of cases. If neither parent has the inversion, then it arose in a gamete de novo or because a parental ovary or testis is mosaic. The specific effects of an inverted chromosome may depend on which genes the flip disrupts. Consulting the “reference” human genome (a consensus of common sequences compiled from many genomes) can help to identify genes that might beimplicated in a specific inversion. Like a translocation carrier, an adult who is heterozygous for an inversion can be healthy, but have reproductive problems. One woman had an inversion in the long arm of chromosome 15 and had two spontaneous abortions, two stillbirths, and two children who died within days of birth. She did eventually give birth to a healthy child. How did the inversion cause these problems? Inversions with such devastating effects can be traced to meiosis, when a crossover occurs between the inverted chromosome segment and the non inverted homolog. To allow the genes to align, the inverted chromosome forms a loop. When crossovers occur within the loop, some areas are duplicated and some deleted in the resulting recombinant chromosomes. In inversions, the atypical chromosomes result from the chromatids that crossed over. Two types of inversions are distinguished by the position of the centromere relative to the inverted section. Figure 13.19 shows a paracentric inversion, in which the inverted section does not include the centromere. A single crossover within the inverted segment gives rise to one normal, one inversion, and two highly atypical chromatids. One abnormal chromatid retains both centromeres and is termed dicentric. When the cell divides, the two centromeres are pulled to opposite sides of the cell, and the chromatid breaks, leaving pieces with extra or missing segments. The second type of atypical chromatid resulting from a crossover within an inversion loop is a small piece that lacks a centromere, called an acentric fragment. When the cell divides, the fragment is lost because a centromere is required for cell division. A pericentric inversion includes the centromere within the loop. A crossover in the inversion loop produces two chromatids that have duplications and deletions, but one centromere each, plus one normal and one inversion chromatid (figure 13.20). Isochromosomes and Ring Chromosomes An isochromosome is the result of another meiotic error that leads to unbalanced genetic material. It is a chromosome that has identical arms. An isochromosome forms when, during division, the centromeres part in the wrong plane (figure 13.21). Isochromosomes are known for chromosomes 12 and 21 and for the long arms of the X and the Y. For example, a woman may have an isochromosome with the long arm of the X chromosome duplicated but the short arm absent. Chromosomes shaped like rings form in 1 of 25,000 conceptions. Ring chromosomes may arise when telomeres are lost, leaving sticky ends that adhere (figure 13.22). Exposure to radiation can form rings. Any chromosome can circularize. Most ring chromosomes consist of DNA repeats and do not affect health. The few that do arise de novo. They all cause a small head (microcephaly), learning disabilities or intellectual disabilities, slow growth, and unusual facial features. 13.5 Uniparental Disomy-A Double Dose from One Parent If nondisjunction (unequal chromosome division) occurs in both a sperm and an oocyte that join, a pair of chromosomes (or their parts) can come solely from one parent, rather than one from each parent, as Mendel’s law of segregation predicts. For example, if a sperm that does not have a chromosome 14 fertilizes an ovum that has two copies of that chromosome, then an individual with the normal 46 chromosomes results, but both copies of chromosome 14 come from the mother. Inheriting two chromosomes or chromosome segments from one parent is called uniparental disomy (UPD) (“two bodies from one parent”). UPD can also arise from a trisomic embryo in which some cells lose the extra chromosome, leaving two homologs from one parent. For example, an embryo may have trisomy 21, with the extra chromosome 21 coming from the father. If in some cells the chromosome 21 from the mother is lost, then both remaining copies of the chromosome are from the father. Because UPD requires the simultaneous occurrence of two rare events—either nondisjunction of the same chromosome in sperm and oocyte or trisomy followed by chromosome loss—it is indeed rare. In addition, many cases of UPD are probably never seen, because bringing together identical homologs inherited from one parent could give the fertilized ovum homozygous lethal alleles. Development would halt. Other cases of UPD may go undetected if they cause known recessive conditions and both parents are assumed to be carriers, when only one parent contributed to the offspring’s illness. This situation was how UPD was discovered. In 1988, Arthur Beaudet of the Baylor College of Medicine saw an unusual patient with cystic fibrosis (see Clinical Connection 4.1). In comparing CFTR alleles of the patient to those of her parents, Beaudet found that only the mother was a carrier—the father had two normal alleles. Beaudet constructed haplotypes for each parent’s chromosome 7, which includes the CFTR gene, and he found that the daughter had two copies from her mother, and none from her father (figure 13.23). How did this happen? Apparently, in the patient’s mother, nondisjunction of chromosome 7 in meiosis II led to formation of an oocyte bearing two identical copies of the chromosome, instead of the usual one copy. A sperm that had also undergone nondisjunction and did not have a chromosome 7 then fertilized the abnormal oocyte. The mother’s extra genetic material compensated for the father’s deficit, but unfortunately, the child inherited a double dose of the mother’s chromosome that carried the mutant CFTR allele. In effect, inheriting two of the same chromosome from one parent shatters the protection that combining genetic material from two individuals offers. This protection is the defining characteristic of sexual reproduction. UPD may also cause disease if it removes the contribution of the important parent for an imprinted gene. Recall from chapter 6 that an imprinted gene is expressed if it comes from one parent, but silenced if it comes from the other (see figure 6.12). If UPD removes the parental genetic material that must be present for a critical gene to be expressed, a mutant phenotype results. The classic example of UPD disrupting imprinting is Prader-Willi syndrome and Angelman syndrome, for which UPD causes 20 to 30 percent of cases (see figure 6.13). These diseases arise from mutations in different genes that are closely linked in a region of the long arm of chromosome 15, where imprinting occurs. They both cause intellectual disability and a variety of other symptoms, but are quite distinct. Some children with Prader-Willi syndrome have two parts of the long arm of chromosome 15 from their mothers. The disease results because the father’s Prader-Willi gene must be expressed for the child to not have the associated illness. For Angelman syndrome, the situation is reversed. Children have a double dose of their father’s DNA in the same chromosomal region implicated in Prader-Willi syndrome, with no maternal contribution. The mother’s gene must be present for health. People usually learn their chromosomal makeup only when something goes wrong—when they have a family history of reproductive problems, exposure to a toxin, cancer, or symptoms of a known chromosomal disease. While researchers analyze human genome sequences, chromosome studies will continue to be part of health care. CHAPTER 14: GENOMES 14.1 From Genetics to Genomics Genetics is a young science, genomics younger still. As one field has evolved and led to the other, driven by clever technologies, milestones have come at oddly regular intervals. A century after Gregor Mendel announced and published his findings, the genetic code was deciphered; a century after his laws were rediscovered, the first human genomes were sequenced. Geneticist H. Winkler coined the term genome in 1920. A hybrid of “gene” and “chromosome,” genome then denoted a complete set of chromosomes and genes. The modern definition refers to all the DNA in a haploid set of chromosomes. The term genomics, credited to T. H. Roderick in 1986, indicates the study of genomes. Thoughts of sequencing genomes echoed through much of the twentieth century, as researchers described the units of inheritance from several different perspectives, as this book has done so far. As the twenty-first century dawned, people began having their genomes sequenced. First came two superstars in genetics, J. Craig Venter and James Watson. For a time papers were published adding different ethnic groups to the roster of the sequenced. Then a few wealthy actors, musicians, politicians, scientists, and entrepreneurs funded their personal genome sequencing, interested in diseases in their families or themselves. Curious journalists did and still do have their genomes sequenced and then report on their collection of gene variants in books, blogs, podcasts, TED talks, and articles. People long dead have had their genomes sequenced, including Egyptian mummies, Ötzi the Ice Man, and Neanderthals and Denisovans (ancient humans described in chapter 17). At the same time that a few prominent and representative individuals were having their genomes sequenced, the first large-scale genome sequencing research projects began, with an ultimate goal of incorporating genome information into electronic health records and routine health care. Figure 1.2 shows how the cost of genome sequencing has dropped, from an initial $1 billion to under $1,000, not including the all-important interpretation. More than 100 million people have already had their genomes sequenced. The idea of sequencing a human genome was once nearly unfathomable in its complexity; soon it will likely be routine, and is being done on the earliest embryos to the oldest old. Genome sequence information can already be stored on smartphones. Our genomes are nearly identical, yet with enough variability—we differ at millions of single-base sites—to make life and society interesting. Beginnings in Linkage and Positional Cloning Studies Sequencing what was thought of as “the” human genome unofficially began in the 1980s with deciphering signposts along the chromosomes. Many of the initial steps and tools grew from existing technology. Linkage maps and studies of rare families that had chromosome abnormalities and specific syndromes enabled researchers to assign a few genes to their chromosomes. Automated DNA sequencing took genetic analysis to a new level—information in the form of a living language. The evolution of increasingly detailed genetic maps is similar to zooming in on a geographical satellite map (figure 14.1). A cytogenetic map (of a chromosome) is like a map of California within a map of the United States, highlighting only the largest cities. A linkage map is like a map that depicts the smaller cities and large towns, and a physical map is similar to a geographical map indicating all towns in an area. Finally, a sequence map is the equivalent of a Google map showing all of the buildings in a specific town. Before the first human genomes were sequenced, researchers matched single genes to specific diseases using an approach called positional cloning. The technique began with examining a particular phenotype corresponding to a Mendelian disease in large families. The phenotype was easily matched to a chromosome segment if all the affected individuals shared a chromosome abnormality that relatives without the phenotype did not have. But abnormal chromosomes are rare. Another way that positional cloning located medically important genes on chromosomes used linkage maps that showed parts of a chromosome shared by only the individuals in a family who had the same syndrome. Then researchers isolated pieces of the implicated chromosome, identified short DNA sequences corresponding to the region of interest, and overlapped the pieces, to gradually find the gene behind a phenotype. Positional cloning was an indirect method that narrowed a gene’s chromosomal locus without necessarily knowing the protein product. Throughout the 1980s and 1990s, the method enabled researchers to discover the genes behind Duchenne muscular dystrophy, cystic fibrosis, Huntington disease, and many other single-gene conditions. Positional cloning was slow. It took a decade to go from discovery of a linked genetic marker for Huntington disease to discovery of the gene. Sequencing the First Human Genomes The idea to sequence the first human genome occurred to many researchers at about the same time, but they had different goals. It was first brought up at a meeting held by the Department of Energy (DOE) in 1984 to discuss the long-term population genetic effects of exposure to low-level radiation. In 1985, researchers convening at the University of California, Santa Cruz, called for an institute to sequence the human genome, because sequencing of viral genomes had shown that it could be done. The next year, virologist Renato Dulbecco proposed that the key to understanding the origin of cancer lay in knowing the human genome sequence. Later that year, scientists packed a room at the Cold Spring Harbor Laboratory on New York’s Long Island to discuss the feasibility of a project to sequence the human genome. At first those against the project outnumbered those for it five to one. The major fear was the shifting of goals of life science research from inquiry-based experimentation to amassing huge amounts of data—ironic considering the importance of bioinformatics today. A furious debate ensued. Detractors claimed that the project would be more gruntwork than a creative intellectual endeavor, comparing it to conquering Mt. Everest just because it is there. Practical benefits would be far in the future. Some researchers feared that such a “big science” project would divert government funds from basic research and developing treatments for the still-new AIDS epidemic. Finally, the National Academy of Sciences convened a committee representing both sides to debate the feasibility, risks, and benefits of the project. The naysayers were swayed to the other side. In 1988, Congress authorized the National Institutes of Health (NIH) and the DOE to fund the $3 billion, 15-year Human Genome Project, which began in 1990 with James Watson at the helm. The project set aside 3 percent of its budget for the Ethical, Legal and Social Implications (ELSI) Research Program. It has helped ensure that genetic information is not used to discriminate. Eventually, an international consortium as well as a private company, Celera Genomics, sequenced the first human genomes. The groups worked separately, and one effort finished a few months before the other, but the accomplishment is referred to historically as “the human genome project.” A series of technological improvements sped the genome project. In 1991, a shortcut called expressed sequence tag (EST) technology enabled researchers to quickly pick out genes most likely to be implicated in disease. This was a foreshadowing of future efforts to focus on the exome, the part that encodes protein and is responsible for 85 percent of single-gene diseases. ESTs are short DNA molecules, called complementary or cDNAs, that are made from the mRNAs in a cell type that is abnormal in a particular illness, such as an airway lining cell in cystic fibrosis. ESTs therefore represent gene expression. Also in 1991, researchers began using devices called DNA microarrays (or chips) to display short DNA molecules that bind their DNA or mRNA complements. Microarray technology became important in DNA sequencing (tiling arrays) as well as in assessing gene expression (expression arrays). Computer algorithms were developed to assemble many short pieces of DNA with overlapping end sequences into longer sequences (figure 14.2). When the project began, researchers cut several genomes’ worth of DNA into overlapping pieces of about 40,000 bases (40 kilobases), then randomly cut the pieces into small fragments. The greater the number of overlaps, the more complete the final assembled sequence. The sites of overlap had to be unique sequences, found in only one place in the genome. Overlaps of repeated sequences found in several places in the genome could lead to more than one derived overall sequence—a little like searching a document for the word “that” versus searching for an unusual word, such as “dandelion.” Searching for “dandelion” is more likely to lead to a specific part of a document, whereas “that” may occur in several places—just like repeats in a genome. The use of unique sequences is why the human genome project did not uncover copy number variants. For example, the sequence CTACTACTA would appear only as CTA. Researchers did not at first appreciate the fact that repeats are a different form of information and source of variation than DNA base sequences. A balance was necessary between using DNA pieces large enough to be unique, but not so large that the sequencing would take a very long time. Two general approaches were used to build the long DNA sequences to initially derive the sequence of the human genome (figure 14.3). The “clone-by-clone” technique the U.S. government funded group used aligned DNA pieces one chromosome at a time. The “whole-genome shotgun” approach Celera Genomics used shattered entire genomes, then used an algorithm to identify and align overlaps in a continuous sequence. Whole-genome shotgun sequencing can be compared to cutting the binding off a large book, throwing it into the air and freeing every page, and reassembling book would divide it into bound chapters. Whole-genome shotgunning is faster, but it misses some sections (particularly repeats) that the clone-by-clone method detects. Technical advances continued. In 1995, DNA sequencing was automated, and software was developed that could rapidly locate the unique sequence overlaps among many small pieces of DNA and assemble them, eliminating the need to gather large guidepost pieces. In 1999, the race to sequence the human genome became intensely competitive. The battling factions finally called a truce. On June 26, 2000, J. Craig Venter from Celera Genomics and Francis Collins, representing the International Consortium, flanked President Clinton in the White House Rose Garden to unveil the “first draft” of the human genome sequence. The milestone capped a decade-long project involving thousands of researchers, culminating a century of discovery. The historic June 26 date came about because it was the only opening on the White House calendar! In other words, the work was monumental; its announcement, somewhat staged. Figure 14.4 is a conceptual overview of genome sequencing. Types of Information in Human Genomes The human genome sequenced by the Public Consortium was actually a composite of the genomes of different individuals. The first two genomes from specific individuals to be sequenced and the findings published, of genome research pioneers J. Craig Venter and James Watson, yielded few surprises. Instead, they showed that we had greatly underestimated genetic variation by focusing only on the DNA base sequence. The numbers of copies of short sequences—copy number variants, or CNVs—contribute significantly to genetic variation, too. Venter learned that he has gene variants associated with increased risk of Alzheimer disease and cardiovascular disease. His genome sequence confirmed that he has genotypes that cause or are associated with dry earwax, blue eyes, lactose intolerance, a preference for activities in the evening, and tendencies toward antisocial behavior, novelty seeking, and substance abuse. He metabolizes caffeine fast, which he also knew. James Watson, according to his genome sequence, carries a dozen rare recessive diseases that would affect glycogen storage, vision, and DNA repair if homozygous, and he is at elevated risk for 20 other diseases. He elected not to learn his status for the ApoE4 gene variant that increases risk of Alzheimer disease, which a grandparent had, but people inferred the result from the surrounding published parts of his genome (a deduction called imputation that compares the surrounding sequence to that of other people’s genomes). He is a slow metabolizer of beta blockers and antipsychotic medications, indicating that he could overdose on normal weight-based dosages of these medications. The third person to have his genome sequenced was called, simply, “YH.” He is Han Chinese, an East Asian population that accounts for 30 percent of modern humanity. He has no inherited diseases in his family, but his genome includes 116 gene variants that cause recessive diseases, as well as many risk alleles. He shares with J. Craig Venter a tendency to tobacco addiction and high-risk alleles for Alzheimer disease. An overall comparison of the first three genome sequences of individuals provided an initial peek at our variation. Each man has about 1.2 million SNPs, but a unique collection. Each has only 0.20 to 0.23 percent of SNPs that are nonsynonymous, meaning that they alter an encoded amino acid, and the men share only 37 percent of these more meaningful SNPs. The math indicates, therefore, that about 0.07 percent of our SNPs may affect our phenotypes. After the first three human genomes were sequenced, others from different population groups were sequenced. Then came the genomes of people who could afford the initially high cost: journalists who were paid to write about the experience and celebrities. Reasons varied, as they still do. The late Steve Jobs (founder of Apple) and late journalist Christopher Hitchens had their cancer genomes sequenced to guide drug choices. In the long run, the genome information did not save their lives. Scholar, documentarian, and host of Finding Your Roots Henry Louis Gates Jr. had his genome sequenced to trace his African origins. An actress and a rock star reportedly did it to better understand mental illness in their families. One geneticist had his genome sequenced to serve as a control for a project to sequence the genomes of all the citizens of Qatar. He discovered that he has gene variants for baldness (which he knew from looking in the mirror), a recessive disease that affects children, Viking ancestors, and most important, a blood clotting disease that explained why he bleeds profusely when injured. Despite learning interesting information, the researcher voiced fears: his family learning things they didn’t want to know and even someone using his DNA sequence to frame him for a crime. 14.2 Analysis of Human Genome Content Before the first drafts of the human genome were completed researchers were already beginning the analysis stage by identifying sites of variation and discovering function of individual genes. Even after hundreds of thousands of human genomes had been sequenced, each new one still revealed nearly 9,000 novel gene variants! To ease sequence comparisons and interpretations, researchers derive a reference genome sequence, which is a digital DNA sequence assembled from the most common base at each point in many sequenced genomes. It is haploid (one copy). Comparing reference genome sequences for different types of organisms and identifying DNA sequences that they share provides views of evolution, discussed in chapter 17. Improving Speed and Coverage Sequencing the first human genomes took 6 to 8 years; today it can be done in hours. Improvements in sequencing technology enabled researchers to work with many more copies of an individual’s genome, which is termed coverage. Recall from section 9.4 that DNA pieces cut from several genomes must be sequenced and overlapped to derive the overall sequence. Because some pieces are lost, the more copies of a genome used, the more likely the overlapping will pick up every base. At least 28 human genome copies are necessary to ensure that most sequences are represented in the final derived sequence. A genome with 40-fold coverage, for example, means each site in the genome is read on average 40 times. High coverage is needed to detect a rare DNA sequence, such as the genome sequencing that helped to diagnose the child described in Clinical Connection 1.1. Even today, genomes are not completely sequenced, because the technology is not perfect. The first human genome sequences published, in 2001, had about 150,000 small gaps; current versions have only a few hundred tiny missing pieces. Sequencing genomes provides much more information than sequencing exomes, which are the exon (coding) parts of protein encoding genes. Knowing the sequences surrounding the exons can be used to detect “structural variants” such as inversions and reciprocal translocations (see figures 13.15 and 13.18), which flip or move DNA but do not alter the base sequence. Exome sequencing arrays must be designed to distinguish between pairs of SNPs on the same homologous chromosomes (in cis configuration) or on different homologs (in trans configuration). The cis configuration indicates linkage, which is important in predicting transmission of a genotype to offspring. The Ongoing Goal: Annotation Just as a book written in a foreign language is meaningless unless translated, knowing the sequence of a human genome is not useful unless we know what the information means. “Annotation” in linguistics means “a note of explanation or comment added to a text or diagram.” In genomics, annotations are descriptions of what genes do, and what the significance of a particular gene variant is likely to be. Researchers are accomplishing this enormous task by meticulously consulting the published scientific literature, DNA sequence databases for every identified gene, and patterns of SNPs in noncoding regions that might be associated with specific disease risks. Genetic counselors and people with doctorates in genetics are doing the annotation, and are called curators, variant scientists, or curation specialists. Annotation of a gene variant might include: the normal function of the gene; mode of inheritance; genotype (heterozygote, homozygote, compound heterozygote); frequency of a variant in a particular population; and classification as benign, likely benign, variant of uncertain significance, likely pathogenic, or pathogenic. Knowing the frequency of a gene variant in a population is important in determining the five classifications for logical reasons. A variant that is common—which means that many people live with it—is less likely to cause a serious illness than one that is rare. If a third of a population has a gene variant that is associated with hypertension, for example, that gene probably contributes only slightly to the overall risk for this complex trait. Otherwise a third of the population would have severe hypertension. However, the reverse is not true—a rare variant can be harmless, and just unusual. To be most valuable, an annotated genome sequence is compared to as much health and family history information as a person can provide. The old-fashioned pedigree is still an invaluable tool and the starting point for many investigations, for this is the information that most people already know— who has what in the family. Genome annotations are also including microbiome data (see section 2.5), because the genes of the organisms that live in and on us can affect expression of our own genes. Limitations of Genome Sequencing Genome sequencing does not provide a complete picture of health. Limitations are technical and conceptual. In a technical sense, genome sequencing cannot detect repeats without additional types of tests because having multiples of a sequence does not alter the sequence. Such repeats include copy number variants, triplet repeat mutations, and even the complete extra chromosomes of trisomies, as well as deletions. Genome sequencing is also blind to uniparental disomy, the rare situation in which a child inherits two alleles of a gene from one parent (see figure 13.23) rather than one from each. The two copies are identical in DNA sequence, and so “count” only once in the overall sequence. In addition, exons that are buried within highly repeated or extensive introns may not be detected. Genome sequencing also does not include mitochondrial genes. In a conceptual sense, genome information must be interpreted to be useful. This means not just identifying the protein that a gene encodes but deciphering all interactions of genes and the gene networks that they form. Imagine a novel that is “read” one letter at a time, so that it is a string of thousands of copies of 26 letters, rather than a nuanced, coherent story with clues and connections conveyed in words. So, too, is a sequenced but unannotated human genome not Informative. Gene interactions are intricate and complex. Figure 11.4 illustrates just a small sampling of diseases that share genes that have altered expression. One gene’s activity affecting the expression of another can explain why siblings with the same single-gene disease suffer to a different extent. For example, a child with severe spinal muscular atrophy (SMA), in which an abnormal protein shortens axons of motor neurons, may have a brother who also inherits the disease but has a milder case thanks to inheriting a variant of a second gene that extends axons. Computational tools and machine learning are used to sort out networks of interacting genes, sometimes called connectomes or interactomes. What were once regarded as simple epistatic interactions—gene affecting gene—may in reality be the tip of an iceberg of complex networks of genes that influence each other. Finally, epigenetic changes induced by environmental factors provide a layer of information that must be applied to DNA information. These are the influences that are perhaps the most important, because we can act on many of them. Practical Medical Matters Until the acceleration of gene discoveries in the 1990s and the sequencing of the first human genomes in the early 2000s, genetics as a medical specialty was a small field, with a few knowledgeable physicians helping families with rare diseases. With the introduction of consumer genetic testing in 2008, the possibility of testing genes not only for disease causing mutations, but for variants that indicate only risk, was suddenly available to anyone—without requiring medical Expertise. Consumer DNA test offerings are abundant, but for health conditions they are not nearly as complete as the types of tests that a medical professional orders. A consumer genetic test for breast cancer, for example, might test for the three most common mutations in BRCA1 and BRCA2, and not the dozens of other mutations in those genes, nor variants of other genes that elevate breast cancer risk. A breast cancer gene test panel that a health care provider orders, in contrast, might test for variants or even complete sequences of more than 100 genes. Genetic and genomic testing as part of health care must meet certain practical criteria. The most important requirement for a regulated DNA test or treatment is clinical utility. Does benefit outweigh risk? Is it as accurate or effective as an existing, approved test or treatment? Will it help people who cannot use the existing test or treatment? Efficacy must be demonstrated. For example, molecular evidence may indicate that people with a particular genotype might respond better to a certain drug than people with different genotypes. A clinical trial must evaluate the drug in both groups of people to demonstrate that prescribing-by-genotype—pharmacogenetics—is helpful on the whole-person level. A DNA test result alone is not sufficient to diagnose a disease, but may support a clinical diagnosis based on symptoms and the results of other types of tests. For example, a person might be a heterozygote for familial hypercholesterolemia, but is not diagnosed with the condition unless the serum cholesterol level becomes elevated or a cardiovascular condition develops (see figure 5.2). However, knowing that a mutation is present can motivate a person to seek further testing, as is the case for breast cancer.  The uncertainty in genetic and genomic testing that makes further diagnostic testing necessary arises from the complications of Mendel’s laws discussed in chapter 5. A DNA test alone is not sufficient for diagnosis because of: incomplete penetrance (genotype does not always foretell phenotype), variable expressivity (different severities in different individuals), epistasis (gene-gene interactions), genetic heterogeneity (mutation in more than one gene causing a phenotype), and environmental influences (epigenetics). How will electronic medical records handle the nuances of genomic data? Will records include entire sequences of the DNA bases A, T, C, and G, or just the diagnostic report that includes gene variants classified as pathogenic? How will the records embrace future discoveries that impact the stored data or diagnoses? How will diagnostic codes work? While these matters are under discussion, the medical profession has had to catch up to both the profusion of new genetic tests and genomic technologies, and the fact that many patients are knowledgeable about DNA—in general and often their own. Genomics is becoming incorporated into more medical specialties, and it already is part of oncology and pediatrics. Medical students analyze their own genomes, and physicians are attending continuing medical education programs to learn genomics. Clinical geneticists, genetic counselors, and molecular pathologists are the specialists who are leading the way in the new genomic medicine. While many researchers are busy annotating human genomes, which are highly complex, others are using chemical and microbiological methods to synthesize simple genomes. The effort is part of a field called synthetic biology. Bioethics discusses synthetic genomes. 14.3 A Genomic View Expands Knowledge The ability to sequence human genomes enables researchers to look backward and add details to what we have known about our genetics, and also to look forward to how the expression of gene variants identified at birth will affect health over a lifetime. Different investigations have sequenced genomes to differing degrees. Exome sequencing refers to the exons (the protein encoding parts), genomic sequencing to more than an exome but less than a full genome, and whole genome sequencing to entire genomes derived from many overlapped copies. Research at the genome level also includes genome-wide association studies, which consult sets of single nucleotide polymorphisms (SNPs) dispersed among the chromosomes (see figures 7.11 and 7.12). Here are two examples of applications of genome information. A New View of Crossing Over Genome sequencing has enabled researchers to look closely at exactly what happens to the region of a pair of chromosomes that cross over during prophase of the first meiotic division, as oocytes and sperm form. Figures 3.5 and 5.11 depict crossing over between homologous chromosomes with a color change of blue to pink. Genome sequencing not only confirms that crossing over occurs, but also reveals instability in the neighboring regions along the involved chromosomes. Researchers analyzed genome sequences from 155,250 people going back several generations from a database that includes health information from most of the population of Iceland. The country has been reproductively isolated for many years, which is why genetic studies have been able to identify many of their disease-causing gene variants. The Icelandic maps of genetic markers were crucial in establishing the first human reference genomes in the early 2000s, and remain important today.  The Icelandic genomes enabled researchers an unprecedented view of two processes that foster human genetic diversity: crossing over and generation of de novo (new) mutations. By comparing DNA sequences at various points along the chromosomes among many people, the researchers uncovered the sites of 4.5 million crossover events that recombined parental chromosomes. They also identified more than 200,000 de novo mutations. Epigenetic changes, such as the patterns with which histone proteins bind to DNA, expose the areas of chromosomes where crossing over occurs (see figure 11.6). The findings reveal that both crossing over and generation of de novo mutations are not as random as had been thought: at least 35 genes affect crossover frequency and location. In addition, the two processes—crossing over and de novo mutations—are connected: mutations are more than 50 times more likely to arise at the sites of crossovers than elsewhere in the genome (figure 14.5). The sexes differ, which reflects epigenetics. More crossovers occur within one kilobase (1,000 bases) of the crossover site in both sexes, but female genomes continue to cross over at higher frequencies than elsewhere up to 40 kilobases from the crossover site. The frequency of crossing over increases with maternal age. However, the genomes of males are more likely to have the de novo mutations that are a major source of single-gene diseases in children. These findings confirm the observations that sperm tend to introduce single-gene mutations whereas oocytes introduce chromosome imbalances, as well as the association of advanced maternal age with chromosomal anomalies in offspring (see figure 13.6). One way to interpret these findings is to take a broad, evolutionary view. The diversity that our genomes naturally generate promotes success of humans as a species, but at the expense of recessive alleles that cause genetic diseases when paired in an individual. Sequencing the Genomes of Newborns Knowing the genome sequence of an individual from birth can have medical advantages. These may include earlier diagnosis of childhood diseases, identifying conditions that might be prevented with early intervention, and providing a backdrop of information for the electronic health record that will follow a person for life. However, obtaining genome information from birth also raises issues of use or abuse in a variety of scenarios. To explore the feasibility, value, and possible risks of newborn genome sequencing, the U.S. government funded a pilot study called BabySeq. Two Boston-area hospitals enrolled 127 healthy newborns and 32 babies in the neonatal intensive care unit (NICU). Half of the children in each group had their genomes sequenced. Those who didn't have their genomes sequenced gave a family history and had standard newborn screening for a few dozen genetic diseases from a blood sample taken from the heel. The researchers and the parents couldn’t tell which babies had their genomes sequenced and which had not. The results attest to the potential value of newborn genome sequencing. The intervention detected childhood onset diseases that hadn’t yet caused symptoms in 15 (9.4 percent) of the babies. Ten of them were from the healthy cohort and five from the NICU group, but none of the mutations were the reasons that the children were in the NICU. Of the 159 infants, 140 were found to have at least one recessive allele that would cause a disease if present in two copies. This type of information would alert individuals, albeit very early, to have prenatal testing when considering parenthood.