Proc. Nad. Acad. Sci. USA Vol. 81, pp. 7875-7879, December 1984 Genetics Identification of a recent recombination event within the human (3-globin gene cluster (meiosis/genetic linkage/chromosome 11/haplotypes) DANIELA S. GERHARD*, KENNETH K. KIDDt, JUDITH R. KIDDt, JANICE A. EGELANDt, AND DAVID E. HOUSMAN* *Center for Cancer Research, Massachusetts Institute of Technology, Cambridge, MA 02139; tDepartment of Human Genetics, Yale University School of Medicine, New Haven, CT 06511; and tDepartment of Psychiatry, University of Miami, Miami, FL 33101 Communicated by Edward A. Adelberg, August 22, 1984 ABSTRACT In a detailed study of inheritance of DNA se- quence polymorphism in a large reference pedigree, an indi- vidual was identified with an apparent genetic recombination event within the human j3globin gene duster. Analysis of the haplotypes of relevant individuals within this pedigree suggest- ed that the meiotic crossing-over event is likely to have oc- curred within a 19.8-kilobase-pair region of the fglobin gene cluster. Analysis of other DNA markers closely linked to the (3- globin gene cluster-segment 12 of chromosome 11 (D11S12) and loci for insulin, the cellular oncogene c-Ha-ras, and pre- proparathyroid hormone-confirmed that a crossover event must have occurred within the region of chromosome 11 be- tween D11S12 and the (-globin gene cluster. It is suggested that the event observed has occurred within a DNA region compatible with recombinational "hot spots" suggested by population studies. Genetic recombination in mammals is a process that has been under study for many decades. Attempts to identify the molecular basis of this process have been hindered by the high complexity of mammalian genomes. To precisely ana- lyze meiotic recombination mechanisms, it would be ex- tremely desirable to identify the DNA segments that recently have undergone crossing-over at meiosis. To achieve this goal, it is necessary to have available a significant number of genetic markers distributed over a chromosomal DNA seg- ment of 100 kilobase pairs (kbp) or less. At present, howev- er, most genetic markers are distributed at such great dis- tances along a mammalian chromosome that precise delinea- tion of the site of a crossover is quite difficult. One region of the human genome that is relatively favorable for a detailed analysis is the ,3globin gene cluster (HBBC) of human chro- mosome 11. Within a 63-kbp region starting 4.2 kbp 5' of the e-globin gene and extending 19 kbp beyond the 3' end of the ,B-globin gene, there are at least 17 polymorphic restriction enzyme recognition sites (1). The question of whether recombination occurs with equal frequency at all sites along the chromosome is one of signifi- cance in understanding the molecular basis of meiotic re- combination. Evidence suggesting that frequency of recom- bination is not evenly distributed within this region has been obtained through population studies of these polymorphic loci. Antonarakis, Orkin, and their collaborators (2, 3) have analyzed HBBC haplotypes in various human populations and demonstrated significant linkage disequilibrium among some alleles within the cluster. Their data suggest that re- combination events are not evenly distributed within the re- gion. These authors propose that crossovers occur with a relatively high frequency within a localized region of 11 kbp 5' of the f3globin gene. In the course of linkage studies of a large multigenera- tional pedigree, we analyzed the segregation of several re- striction fragment length polymorphisms (RFLPs) on the short arm of chromosome 11, including several in the 3-glo- bin gene region. This analysis led to the identification of an individual with one chromosome that appears to be a prod- uct of a recombination within the l3-globin gene region itself. Since detailed analysis of the structure of such recombinant chromosomes should give insight into the distribution of and mechanisms underlying meiotic crossing-over, we report here our initial analyses of this event. MATERIALS AND METHODS Pedigree Studied. The pedigree reported on here is part of the Old Order Amish pedigree no. 110 previously studied by Egeland and collaborators (4, 5). Lymphoblast and fibro- blast cell lines have been established on 51 individuals in this kindred by the Institute for Medical Research (Camden, NJ). Preparations of DNA. DNA was extracted from cultured cells by standard procedures (6). Analysis of DNA Polymorphisms. DNAs were digested with different restriction enzymes that have been shown to identify polymorphic sites when hybridized with a given probe on a Southern blot. The digests were performed with 2-3 units/pg of DNA overnight under conditions specified by the manufacturer. DNAs were run on appropriate-per- centage gel, transferred as described by Southern (7) to ny- lon-based filter, and probed with nick-translated probes (8). Definitions of Loci Studied. Globin haplotypes for each in- dividual were characterized by the use of the following re- striction enzymes in the Southern hybridization procedure. The HBBC sites were: (i) the HincII site 5' to the E-globin gene (2); (ii) the HindIII sites present in the intervening se- quence 2 (IVS2) of both G% and Ayglobin genes (9, 10); (iii) the HincII sites present in the q,p1-globin gene and 3' to the &81-globin gene (2); (iv) the Taq I site 5' to the 8-globin gene called "E" (11); (v) the Hinfl site 5' to the f3globin gene (12); (vi) the Rsa I site 5' to the ,B-globin gene (27); (vii) the HgiAI site in the first exon of the fglobin gene (3); (viii) the Ava II site in the IVS2 of the fBglobin gene (2); (ix) the HindIII site 3' to the j3globin gene homologous to probe pRK-29 isolated by R. Kaufman as described in ref. 13; and (x) the BamHI site 3' to the fBglobin gene (14). The four other marker loci used in the study are HRASI (human oncogene c-Ha-ras-J), D11S12 (segment 12 of chro- mosome 11), INS (insulin gene), and PTH [gene for prepro- parathyroid hormone, which is enzymatically cleaved to parathyroid hormone (PTH)]. The HRASJ locus was typed on DNAs digested with BamHI (15, 16) and is characterized Abbreviations: INS, insulin locus; HRASI, designation for human locus of oncogene c-Ha-ras-J; PTH, designation for locus of prepro- parathyroid hormone; IVS, intervening sequence; cM, centimor- gans; kbp, kilobase pairs; HBBC, f3globin gene cluster. 7875 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. ยง1734 solely to indicate this fact. D o w n lo a d e d a t C a rn e g ie M e llo n U n iv e rs ity o n A p ri l 5 , 2 0 2 1 Proc. NatL. Acad Sci USA 81 (1984) y C Ar E I29 6 o E lt 0 +---- + -A - +-++ ++Q -+-+++ +- F +_+ - A +---- + + B -++-+ +-D -++-++ - + E - ++ -+ + - + E I~ ~ .I I I . I I i A A C C 6/ A A C D +___-- + - A -+- ++ + + + C 4EFbbti0Eb E E F E F E E' A'AE A' E' A' +---- + - E -++-+ + - + C A C C C A A- E E E C A' A A T IR E ~'R + _ __+ + + .___+ - FIG. 1. Globin haplotypes of key members of the sibship. Partial globin haplotypes are shown for each relevant member. +, Presence of a restriction enzyme recognition site; -, absence of the restriction enzyme recognition site. Individual X exhibits a unique fglobin haplotype not exhibited by her parents, individuals V and W. The f3-globin haplotypes of V and W can be inferred by the /3-globin haplotypes of their parents, individuals R and S and T and W, respectively. Intact transmission of the f3globin gene region of the parents of X to the remaining siblings was observed and indicated by letter designation in Table 1. The designation 8 refers to the BamHI polymorphism 3' to the j-globin structural gene; all other sites near the f-globin gene were not segregating in the relevant individual. by variously sized insertion sequences. Three alleles were distinguishable. The D11S12 locus was typed on Taq I-di- gested DNA; the two alleles were defined by presence or absence of the polymorphic Taq I site (17). PTH was typed on Pst I-digested DNA; two alleles were defined by presence or absence of the polymorphic Pst I site in the third exon (18). INS was studied with both Sac I and Pvu II. The allelic definitions for this restriction fragment length polymorphism are discussed below with the results and discussion. The es- timated map distances of these markers from the globin clus- ter are 9 centimorgans (cM) to INS, 10 cM to HRASI, 3 cM to D11S12, and 7-20 cM to PTH (18-21) (see Fig. 3). RESULTS Fig. 1 illustrates the family structure for the sibship in which the 83-globin haplotype of interest was observed. A possible explanation for an inconsistent globin haplotype would be nonpaternity for individual X; however, 32 previously typed polymorphic marker systems, including HLA, (refs. 5 and 22; unpublished results) are in complete agreement with the biological relationships shown in Fig. 1. Based on the mark- ers alone, the odds of the identified parentage to a random paternal gamete for individual X exceed 3.5 x 104. All other DNA polymorphisms studied were consistent with parental genotypes, further increasing the odds that the attributed pa- ternity is correct. These results, which led us to reject non- G AYr Hloc a 10 kb paternity as an explanation for the unexpected HBBC haplo- type in individual X, are consistent with previous studies of the general nature of this community (23, 24) and personal knowledge (J.A.E.) of this cohesive family unit. The availability of all four grandparents of the core sibship has made the establishment of linkage phase in the two par- ents (V and W in Fig. 1) unequivocal for the markers in the HBBC as well as for the other loci studied. Six different HBBC haplotypes are observed segregating in this pedigree. Polymorphic loci within the HBBC are depicted in Fig. 2. Eight of the 12 polymorphic sites shown in Fig. 2 were used to establish the globin haplotype shown in Fig. 1. Haplo- types A and E are each represented twice in the grandpar- ents. The two parents (V and W in Fig. 1) have the genotypes A/C and A'/E, respectively. These genotypes predict four possible types of offspring: namely, AA', A'C, CE, and A'E. All four combinations are represented; two of the offspring were A'C, four were AA', one was AE, and four were CE. Individual X has a genotype that could not have arisen by transmission of the intact /3globin gene region carried by the parental chromosomes from her father or her mother. Several explanations appeared initially compatible with the data: (i) a recombination within the P3-globin gene region, (ii) a gene conversion event within the ,3globin gene region, (iii) a new mutation at the BamHI site in the germ cell of either individual X's father or mother, and (iv) an artifact due *A1 8 Hind I |inI Hinft l it Borm HI ToqI HgiAI Hindm ZEN FIG. 2. Map of the HBBC. Arrows indicate the polymorphic restriction enzyme recognition sites._, Coding regions; r, 3-globin pseudo- gene. 7876 Genetics: Gerhard et aL D o w n lo a d e d a t C a rn e g ie M e llo n U n iv e rs ity o n A p ri l 5 , 2 0 2 1 Proc. NatL. Acad Sci USA 81 (1984) 7877 a PTH *- f A I 7-20cM D1192 NOWM INS HRASI 9-10cm 1 b FIG. 3. (a) Order of the five loci studied as determined by linkage analysis. Recombination distance between markers on the short arm of chromosome 11 was estimated by two-point analysis (refs. 19 and 20; unpublished results). The heavy dot represents the centromere. (b) The predicted orientation of the genes within the HBBC with respect to the centromere. to a mutation at the BamHI site in the cultured lymphoblast cell line from which the DNA of individual X was obtained. To address the fourth possibility, the typing of individual X was repeated on DNA derived from an independent source. A skin biopsy obtained from individual X was used to derive DNA from cultured fibroblasts. The results of this determination were identical to results with DNA derived from cultured lymphoblast cells of individual X; the BamHI typing was heterozygous. During the course of these studies, determination of the key 3-globin polymorphisms were per- formed two to four times to eliminate any possible typing error in the /3globin haplotypes. To distinguish among the remaining three possibilities, we attempted to establish whether the chromosome 11 carrying the + - - - - - - - f-globin haplotype had undergone an ob- ligate crossover within the region of lip carrying the HBBC. Data were obtained for four markers known to be closely linked to the HBBC (18, 19, 21). These markers and their estimated map distance from the HBBC are shown in Fig. 3a. Table 1 gives the genotypes of the relevant individuals in the sibship for these markers. It is possible to determine link- age phase for all of the markers for the two parents in Fig. 1 (V and W, Fig. 4). These assignments are compatible with the genotypes of the 11 other children of this couple (Table 1). Both parents of individual X are heterozygous (+/-) for the DlS12 DNA segment. The genotypes of the grandpar- Table 1. Genotypes of five loci of family members analyzed Genotypes Individual PTH HBBC D11S12 HRASI INS R +/+ A/B +/+ 1/1 3/3 S +/+ CID ND 1/1 1/1 T -/- FIE +/- 3/3 2/9 U +/- A'/E' +/- 1/3 3/3 V +/+ A/C +/- 1/1 1/3 W -/- A'/E +/- 1/3 2/3 X +/- +/+ 1/1 3/3 +/- C/A' +/+ 1/1 3/3 +/- A/A' +/- 1/1 1/3 +/- C/E +/- 1/3 2/3 +/- C/E -/- 1/3 1/2 ND C/E +/- 1/3 2/3 +/- C/E +/- 1/3 2/3 +/- A/A' +/- 1/1 ND ND A/E -/- 1/3 1/2 +/- A/A' +/- 1/3 1/2 +/- A/A' +/- 1/1 1/3 +/- C/A' +/+ 1/1 3/3 ND, not determined. ents of individual X establish the phase of the D11S12 alleles in the mother of individual X. Individual X's maternal grand- mother (individual U) transmitted HBBC haplotype A' to her daughter. Since U was homozygous for the presence of the Taq I site in D11S12, we must conclude that W carried the + allele for D11S12 and HBBC A' on one chromosome and the - allele for D11S12 and HBBC E on the other chromosome. This assignment is consistent with all typing of the 11 sib- lings of individual X. Individual V, the father of individual X, was also heterozygous at D11S12. Typing of his 11 other offspring establishes that in V the HBBC haplotype A is as- sociated with the absence of the Taq I site while haplotype C is associated with the presence of this Taq I site. Individual X is homozygous for the presence of the Taq I site. The ge- notype of individual X at each of the polymorphic sites of the f3-globin gene region except the 3' BamHI site and at D11S12 requires an obligate paternal crossover between the HBBC and D11S12. In contrast to the other sites in the HBBC, no crossover is required between the 3' BamHI site and D11S12. Even if the BamHI site had been converted from - to + by a gene conversion event, a crossover between D11S12 and the HBBC would still have been required by the data. Further evidence supporting the occurrence of a pater- nal crossover within this region is provided by analysis of genotypes at the HRASJ locus. At the HRASI locus, the father was homozygous for the smallest of the three BamHI fragments-1,1-while the mother was heterozygous, hav- ing one copy of the smallest BamHI fragment and one copy of the largest BamHI fragment-1,3. The haplotypes shown in Fig. 4 support the view that the maternal chromosome lip was transmitted intact without crossover in the region be- tween the HBBC and HRASJ. The most parsimonious ex- PTH c Gr A k18 E 0DIIS12 HRASI INS I , //+ + 1 3I+ -++ -+ + + X - 3 2cd -if + 3 -0......................?2 + - -::::: 4::: ::::,: .....:, .. '. -:- ,-, 1'.'..:.. 7fT11 FIG. 4. Haplotypes of the parents W and V and individual X determined from typings of all kindreds in this sibship. +, Restric- tion enzyme site is present; -, restriction enzyme site is absent. The typings were done on DNA digested with enzymes indicated in Ma- terials and Methods. HRAS1 allele I is a 6.6-kbp BamHI fragment, while allele 3 is 8.4 kbp. The three alleles of INS are: 1, 800 bp; 2, 830 bp; and 3, 900 bp. Genetics: Gerhard et aL ...... D o w n lo a d e d a t C a rn e g ie M e llo n U n iv e rs ity o n A p ri l 5 , 2 0 2 1 Proc. NatL Acad Scd USA 81 (1984) planation of the genotype of individual X is thus a single pa- ternal crossover event occurring between the Taq I site at the 5' side of the 8-globin gene ("E") and the BamHI site at the 3' side of the 83-globin gene. This interpretation of the data specifies the orientation of the genes within the HBBC with respect to the other markers on lip. If a crossover has occurred within the 13-globin gene region, then the D11S12 DNA sequence must be closest to the 3' end of the HBBC. The data are thus most consistent with the gene order shown in Fig. 3b. If, however, a gene conversion event was responsible for the BamnHI genotype of individual X, then no inference can be made regarding the gene order given in Fig. 3b. The polymorphic site 5' to the INS gene is due to insertion of sequences that are made up of 14-bp repeats (25). In the general population, a wide range of size variation is observed in this region of the DNA. This variation is exhibited by di- gestion of the DNA with Sac I and hybridization to a probe that contains a segment of the 5' region of INS. Within the segment of the Old Order Amish population that we have studied, considerably less variation in length of this DNA segment is observed than in the general population. To clear- ly identify sequence length variation in INS, we used the probe pHins310, which includes the region directly adjacent to the variable DNA segment (26). This identifies DNA frag- ments from this region, which vary in length between =800 bp and =900 bp. Allele I is "800 bp, allele 2 is -830 bp, and allele 3 is -900 bp. The father and the mother of individual X are heterozygous for this locus. These alleles are assigned to the paternal and maternal chromosomes as shown in Fig. 4. The fact that individual X is homozygous for allele 3 of INS also supports the conclusion that a paternal crossover oc- curred between INS and the 5' cluster of sites in the HBBC but that no crossovers occurred between INS and D11S12 or between INS and the BamHI site 3' of the HBBC. Typing of the PTH locus did not give additional informa- tion since the father was homozygous for the presence of the polymorphic Pst I site, while the mother was homozygous for the absence of this site. Individual X is, as expected, het- erozygous for this site. In an attempt to further delineate the site of the crossover, we typed relevant individuals in the sibship for a number of reported polymorphisms around and within the HBBC. The HgiAI, Hinfl, Rsa I, and Ava II sites (Fig. 2 and Materials and Methods) proved to be uninformative because both the father and mother of individual X were homozygous for the same allele in each case. DISCUSSION The data presented here demonstrate either a crossover event within the fglobin gene cluster or a gene conversion event encompassing a segment of the cluster. Previous stud- ies based on population genetic methods have suggested that recombination within the cluster may be relatively much more frequent between the qif1- and f-globin genes than in the -20 kbp on either side (2, 3). This conclusion is based upon the observation of nonrandom associations of alleles at the various polymorphic sites. Alleles within the cluster of polymorphic sites stretching from 5' of the e-globin gene to 3' of q4pl-globin pseudogene show strong linkage disequilib- rium; similar linkage disequilibrium exists for the alleles in the cluster of sites from 5' of the f-globin gene to 17 kbp downstream. No linkage disequilibrium exists between these two subclusters. Since recombination is the primary evolu- tionary mechanism for reducing linkage disequilibrium, the implication of the population data is that recombination rate is relatively much higher between the two subclusters than within either subcluster. Thus, on an evolutionary time scale, this region has been implicated as a "hot spot" for recombination (see ref. 1 for review). Chakravarti et al. (27) have used population genetic theory and the observed link- age disequilibripm in the HBBC to estimate the rate of re- combination in the 5' subcluster, in the 3' subcluster, and in the small "hot spot" between them. Their calculations, based upon two-locus linkage disequilibrium theory for sites on opposite sides of the "hot spot," is 0.003. In a previous study, Stamatoyannopoulos et al. (28) estimated the recom- bination between the S-globin structural gene and 3-thalasse- mia to be 0.03. Assuming that our observations represent a single crossover event, we can estimate the recombination rate between the qip- and ,-globin loci. We have typed other branches of this Old Order Amish pedigree and a large Vene- zuelan pedigree (29) for these same loci (19, 20). A modified version of LIPED (30, 31) gave a maximum lod score (loga- rithm of odds of linkage) of 21.49 at =0 0.018 + 0.021; the two-support-unit confidence interval (32) extends from 6 = 0.0037 to 6 = 0.056. Thus, our estimate of the recombination frequency in this interval is not significantly greater than Chakravarti's. All three of these estimates are significantly higher than that predicted on distance in nucleotides alone. We do not have as clear an expectation for the rate of gene conversion. In no other case in our study have we observed an anomalous transmission of a multisite haplotype, but it is difficult to calculate the exact number of nonconversions definitely observed. We can say that, if the 3' BamHI site was converted, we also observed an independent crossover event between the HBBC and D11S12. This second event has a prior probability of only 3% since the map distance involved is 3 cM (19, 20). Although this probability is low, it is not sufficiently low to reject the hypothesis. If we accept the hypothesis that a crossover within the HBBC has been observed, it is of interest to note that the crossover falls within the region suggested to be a hot spot for recombination. The identification of additional polymor- phic sites in the region of interest would clearly be of utility in delineating the site of the crossover. The most direct ap- proach to this problem would be to isolate DNA from the crossover region from the relevant individuals by recombi- nant DNA techniques and to identify DNA sequence poly- morphism directly by DNA sequencing methods. It would be of interest to compare the data obtained here with results in other well-mapped regions of the human genome. The most obvious candidate for such an analysis would be the human major histocompatibility complex (MHC). Crossover individuals have already been identified in this region by immunological methods. Sites of recombi- nation also have been studied in the murine MHC. Data from this region of the mouse genotne indicate that specific hot spots for recombination exist within this region (33). Com- parison of LNA sequences at sites of recombination in the two species would make it possible to determine whether crossover sites within the human and mouse MHC regions exhibit sequence' homology to sites of crossing-over in the HBBC. The data presented here allow a prediction to be made on the orientation of the HBBC with respect to other markers in the adjacent region of chromosome 11. The data presented here are most consistent with the organization of this region of chromosome 11 as shown in Figs. 3b and 4. In this inter- pretation, the E globin gene region of the HBBC is closest to PTH, and the 3globin gene is closest to INS, HRASJ, and D11S12. Note Added in Proof. We have completed the genotype analysis of the restriction fragment length polymorphism associated with the Rsa I site 9.1 kbp 3' of the BamHI site (27). The genotype of individ- ual X is +/- at this Rsa I site Individual V is +/- at this site, while individual W is +/+. Typing of remaining members of the sibship assigns the - allele for individual V to the C haplotype for HBBC. This result is consistent with the occurrence of a crossover 7878 Genetics: Gerhard et aL D o w n lo a d e d a t C a rn e g ie M e llo n U n iv e rs ity o n A p ri l 5 , 2 0 2 1 Proc. NatL Acad Sci USA 81 (1984) 7879 within the region indicated in this paper. If a gene conversion has occurred, it must have extended over a region of at least 9.1 kbp. We wish to thank the following people: H. Kronenberg for pPTHM122, G. Bell for pHins310, C. Tabin for pEJ6.6, R. White for pADJ-762, and S. Antdnarakis, H. Kazazian, R. Kaufman, and B. Forget for the HBBC probes. This work was supported in part by Grants GM 27882 and CA 17575 to D.E.H.; NS 11786, MH 39239, and MH 30909 to K.K.k.; MH 28287 to J.A.E.; and a Damon Run- yon-Walter Winchell Cancer Fund Fellowship DRG-665 to D.S.G. 1. Collins, F. S. & Weissman, S. M. (1984) Prog. Nucleic Acids Res. Mol. Biol. 28, in press. 2. Antonarakis, A. E., Boehm, C. D., Giardina, P J. V. & Kaza- zian, H. H. (1982) Proc. Nail. Acad. Sci. USA 19, 137-141. 3. Orkin, S. H., Kazazian, H. H., Jr., Antonarakis, S. E., Goff, S. C., Boehm, C. D., Sexton, J. P., Waber, P. G. & Giardina, P. J. V. (1982) Nature (London) 296, 622-631. 4. Egeland, J. A., Kidd, J. R., Frazer, A., Kidd, K. K. & Neu- hauser, V. I. (1984) Am. J. Psychiatry 141, 1049-1054. 5. Kidd, K. K., Egeland, J. A., Molthan, L., Pauls, D. L., Kru- ger, S. D. & Messner, K. H. (1984) Am. J. Psychiatry 141, 1042-1048. 6. Gtsella, J., Varsanyi-Breiner, A., Kao, F., Jones, C., Puck, T. T., Keys, C., Orkin, S. & Housman, D. (1979) Proc. Nail. Acad. Sci. USA 76, 5239-5243. 7. Southern, E. M. (1975) J. Mol. Biol. 98, 503-517. 8. Nordstrom, J. L., Roop, D. R., Tsai, M. J. & O'Malley, B. W. (1979) Nature (London) 278, 328-331. 9. Jeffreys, A. J. (1979) Cell 18, 1-10. 10. Tuan, D., Biro, P. A., de Riel, J. K., Lazarus, H. & Forget, B. G. (1979) Nucleic Acids Res. 6, 2519-2544. 11. Maeda, N., Bliska, .. B. & Smithies, 0. (1983) Proc. NatI. Acad. Sci. USA 80, 5012-5016. 12. Moshonas, N., Boer, E. & Flavell, R. A. (1982) Nucleic Acids Res. 10, 2109-2120. 13. Tuan, D., Feingold, E., Newman, M., Weissman, S. M. & Forget, B. G. (1983) Proc. Natl. Acad. Sci. USA 80, 6937- 6941. 14. Kan, Y. W., Lee, K. Y. & Furbetta, M. (1980) N. Engl. J. Med. 302, 185-188. 15. Goldfarb, M., Shimizu, K., Perucho, M. & Wigler, M. (1982) Nature (London) 296, 404-409. 16. Capon, D. J., Chen, E, Y., Levinson, A. D., Seeberg, P. H. & Goeddel, D. V. (1983) Nature (London) 302, 33-37. 17. Naylor, S. L., Sakaguchi, A. Y., Barker, D., White, R. & Shows, T. B. (1984) Proc. Natl. Acad. Sci. USA 81, 2447- 2451. 18. Antonarakis, S. E., Phillips, J. A., Mallonee, R. L., Kazazian, H. H., Fearon, E. R., Waber, P. G., Kronenberg, H. M., Ull- rich, A. & Meyers, D. A. (1983) Proc. Natl. Acad. Sci. USA 80, 6615-6619. 19. Gerhard, D. S., Kidd, K. K., Housman, D., Gusella, J. & Kidd, J. R. (1984) Cytogenet. Cell Genet. 37, Nos. 1-4. 20. Gerhard, D. S., Kidd, K. K., Kidd, J. R., Gusella, J. & Hous- man, D. (1984) ICSU Short Rep. 1, 172-173. 21. Fearon, E. R., Antonarakis, S. E., Meyers, D. A. & Levine, M. A. (1984) Am. J. Hum. Genet. 36, 329-337. 22. Molthan, L., Kidd, K. K. & Egeland, J. E. (1982) Can. J. Med. Technol. 44, 99-101. 23. McKusick, V. A., Hostetler, J. A. & Egeland, J. A. (1964) Bull. Johns Hopkins Hospital 115, 203-222; reprinted in McKusick, V. A. (1978) Medical Genetic Studies ofthe Amish (Johns Hopkins Univ. Press, Baltimore). 24. Cross, H. E. & McKusick, V. A. (1970) Social Biol. 17, 83- 101. 25. Bell, G. I., Karam, J. & Rutter, W. J. (1981) Proc. Natl. Acad. Sci. USA 78, 5759-5763. 26. Bell, G. I., Selby, M. J. & Rutter, W. J. (1982) Nature (Lon- don) 295, 31-34. 27. Chakravarti, A., Buetow, K. H., Antonarakis, S. E., Weber, P. G., Boehm, C. D. & Kazazian, H. H. (1984) Am. J. Hum. Genet. 36, in press. 28. Stamatoyannopolos, G., Weitkamp, L. R., Kotsakis, P. & Ak- rivakis, A. (1977) Hemoglobin 1, 561-570. 29. Wexler, N. S., Bonilla, E., Young, A. B., Shoulson, I., Go- mez, F., Starosta, S., Travers, H., Villalobas, M., de Quiroz, I., Erbe, R., Penney, J. B., Uzzell, R. S., Burnham, F. A., Daugherty, L., Jones, B., Mapstone, C., Rivas, M., Messer, E., Wexler, A., Snodgrass, R., Rosenzweig, G., Esteves, J., Marsol, N., Bailey, S., Brinley, F. J., Goldstein, E., Greene, A. E., Kidd, J. R., Kidd, K. K., Gusella, J. F., Conneally, P. M. & Moreno, H. (1984) Cytogenet. Cell Genet. 37, 605. 30. Ott, J. (1976) Am. J. Hum. Genet. 28, 528-529. 31. Hodge, S. E., Morton, L. A., Tideman, S., Kidd, K. K. & Spencei M. A. (1979) Am. J. Hum. Genet. 31, 761-762. 32. Edwards, A. W. S. (1972) Likelihood (Cambridge Univ. Press, Cambridge, England), p. 275. 33. Steinmetz, M., Minard, K., Horvath, S., McNicholas, J., Fre- linger, J., Wake, C., Long, E., Mach, B. & Hood, L. (1982) Nature (London) 300, 35-42. Genetics: Gerhard et aL D o w n lo a d e d a t C a rn e g ie M e llo n U n iv e rs ity o n A p ri l 5 , 2 0 2 1