A Schizophrenia Locus May Be Located inRegion 10p15–p11

Richard E. Straub,1* Charles J. MacLean,1,2 Rory B. Martin,1 Yunlong Ma,1 Maxim V. Myakishev,1Carole Harris-Kerr,1 Bradley T. Webb,1 F. Anthony O’Neill,3 Dermot Walsh,4 andKenneth S. Kendler1,2

1Department of Psychiatry, Virginia Institute for Psychiatric and Behavioral Genetics, Medical College ofVirginia/Virginia Commonwealth University, Richmond, Virginia

2Department of Human Genetics, Virginia Institute for Psychiatric and Behavioral Genetics, Medical College ofVirginia/Virginia Commonwealth University, Richmond, Virginia

3Department of Psychiatry, The Queens University, Belfast, Northern Ireland4Health Research Board, Dublin, Ireland

In our genomic scan of 265 Irish familieswith schizophrenia, we have thus far gener-ated modest evidence for the presence ofvulnerability genes in three chromosomalregions, i.e., 5q21–q31, 6p24–p22, and 8p22–p21. Outside of those regions, of all markerstested to date, D10S674 produced one of thehighest pairwise heterogeneity lod (H-LOD)scores, 3.2 (P = 0.0004), when initially testedon a subset of 88 families. We then tested atotal of 12 markers across a reagion of 32 cen-timorgans in region 10p15–p11 of all 265families. The strongest evidence for linkageoccurred assuming an intermediate pheno-typic definition, and a recessive geneticmodel. The largest pairwise H-LOD scorewas found with marker D10S2443 (maxi-mum 1.95, P = 0.005). Using multipoint H-LODs, we found a broad peak (maximum1.91, P = 0.006) extending over the 11 centim-organs from marker D10S674 to markerD10S1426. Multipoint nonparametric link-age analysis produced a much broaderpeak, but with the maximum in the same lo-cation near D10S2443 (maximum z = 1.88, P =0.03). Based on estimates from the multi-point analysis, this putative vulnerabilitylocus appears to be segregating in 5–15% ofthe families studied, but this estimateshould be viewed with caution. When evalu-ated in the context of our genome scan re-

sults, the evidence suggests the possibilityof a fourth vulnerability locus for schizo-phrenia in these Irish families, in region10p15–p11. Am. J. Med. Genet. (Neuropsy-chiatr. Genet.) 81:296–301, 1998.© 1998 Wiley-Liss, Inc.

KEY WORDS: schizophrenia; linkage; Irish;chromosome 10

INTRODUCTION

Schizophrenia is a relatively common, often chronic,and debilitating mental illness. Evidence from a vari-ety of studies has clearly demonstrated that geneticfactors contribute substantially to the etiology [Gottes-man and Shields, 1982; Kendler et al., 1993]. For ex-ample, a recent study of a national twin sample fromFinland estimated the heritability of schizophrenia tobe in excess of 80% [Cannon et al., 1998]. Several linesof evidence suggest that the mode of transmission iscomplex, including multiple, possibly interactinggenes, along with genetic heterogeneity and incompletepenetrance [O’Rourke et al., 1982; Kendler and Diehl,1993; Prescott and Gottesman, 1993; Cloninger, 1994].

Given these complexities of the ‘‘genetic architec-ture,’’ in combination with the difficulty of character-ization of the phenotype, it is not too surprising thatwith current sample sizes and analytical methods,genes have yet to be identified. A tremendous amountof work, however, has gone into linkage and associationstudies, and incremental progress is being made [Ka-rayiorgou and Gogos, 1997]. Initial positive linkage re-ports on chromosomes 22q [Pulver et al., 1994], 6p[Wang et al., 1995; Straub et al., 1995], and 8p [Pulveret al., 1995] have been tentatively supported by subse-quent worldwide collaborative efforts [Gill et al., 1996;Schizophrenia Linkage Collaborative Group for Chro-mosomes 3 and 6 and 8, 1996]. More importantly, anumber of complete genome scans are now being fin-

Contract grant sponsor: National Institute of Mental Health;Contract grant numbers: MH41953, MH52537, MH45390.

*Correspondence to: Richard E. Straub, Ph.D., Director, Mo-lecular Genetics Laboratory, Department of Psychiatry, VirginiaInstitute for Psychiatric and Behavioral Genetics, Virginia Com-monwealth University, 800 East Leigh St., Suite 110, Richmond,VA 23219-1534. E-mail: rstraub@hsc.vcu.edu

Received 18 February 1998; Revised 30 March 1998

American Journal of Medical Genetics (Neuropsychiatric Genetics) 81:296–301 (1998)

© 1998 Wiley-Liss, Inc.

ished. A careful and systematic tabulation of the mostpositive regions from these scans (regardless of wheth-er the lod scores from individual markers are deemed‘‘significant’’ or not) should be a productive way to pri-oritize more detailed follow-up studies.

The Irish Study of High-Density SchizophreniaFamilies (ISHDSF) contains 1,408 individuals, withDNA available in 265 families, systematically ascer-tained from psychiatric facilities in Ireland and North-ern Ireland. In previous studies with these families wegenerated support for the presence of three loci influ-encing the vulnerability to schizophrenia on chromo-somes 5q21–q31 [Straub et al., 1997b], 6p24–p22[Straub et al., 1995], and 8p22–p21 [Kendler et al.,1996a]. We employ a split-sample design that allows usto attempt to discriminate true from false positives by‘‘replicating’’ our initial findings internally before com-municating our best results to other investigators. Wetest three equal, randomly divided, and unchangingsubsets (n 4 87 or 88) of families for the first pass scan,rank order the results, and then follow up as many ofthe best regions as possible by testing the same mark-er(s) on additional subsets and by testing flankingmarkers. For our genome scan and follow-up work, wehave tested 620 markers to date: 131 on all 265 fami-lies, 29 markers on two subsets each, and 460 on onesubset each.

Our interest in 10p14–p13 was first raised in May of1997 when marker D10S674 produced a heterogeneitylod H-LOD of 3.2 (P 4 0.0004) on a single subset (n 488) of families, one of the best results we have found.We have followed this observation up with additionalgenotyping and have evaluated the results in the con-text of the distribution of lod scores we have observedacross the genome. We conclude that the results on 10pare relatively noteworthy, and here we report the evi-dence suggesting a possible fourth schizophrenia vul-nerability locus in these Irish families.

MATERIALS AND METHODSPedigree Ascertainment

As detailed elsewhere [Kendler et al., 1996b], theIrish Study of High-Density Schizophrenia Families(ISHDSF) is a collaborative project between the Medi-cal College of Virginia, the Health Research Board,Dublin, and Queen’s University, Belfast. The linkagesample contained 1,408 individuals from 265 system-atically ascertained multiplex schizophrenia familiescollected in Ireland and Northern Ireland. Our entry‘‘field’’ criteria were: two or more first-, second-, orthird-degree relatives who, according to the field psy-chiatrists, met DSM-III-R criteria for schizophrenia[American Psychiatric Association, 1987] or poor out-come schizoaffective disorder (PO-SAD) [Kendler et al.,1995]. No exclusion criteria were used, so families werescreened neither for bilineal descent of schizophreniaspectrum illness, nor for the presence of other forms ofpsychopathology, such as bipolar illness. However, therates of affective illness in relatives from the ISHDSFwere not increased above that found in the relatives ofunselected schizophrenia probands from a case registry[Kendler et al., 1993a]. For inclusion into the linkage

sample, each pedigree had to have two or more first-,second-, or third-degree relatives with a diagnosis ofD1–D5 (see below), one or more of whom had a D1–D2diagnosis; 98.6% of cases diagnosed with schizophreniaor schizoaffective disorder had psychiatric records.

Diagnostic Assessment

The diagnostic instruments included the StructuredInterview for DSM-III-R Diagnosis (SCID) [Spitzer etal., 1992] for selected axis I disorders (psychosis, majordepression, mania, cyclothymia, dysthymia, alcoholabuse/dependence, panic disorder, and generalizedanxiety disorder) and the Structured Interview forSchizotypy (SIS) [Kendler et al., 1989] for putativeschizophrenia spectrum personality disorders (schizo-typal, paranoid, schizoid, and avoidant personality dis-order). Diagnosis was made, using DSM-IIIR criteria,based on all available information (personal history,hospital record, and family history report) by individu-als blind to knowledge of genotypes and to the psycho-pathology of relatives. Final diagnostic review wasdone by K.S.K. and D.W., whose agreement on diag-nostic category was 74% (weighted k: 0.94 ± 0.05). Thediagnostic categories are as follows.

Categories D1–D2 (narrow; 577 affecteds, 285 af-fected sib pairs) included ‘‘core schizophrenic pheno-types:’’ schizophrenia, poor outcome schizoaffective dis-order (PO-SAD), and simple schizophrenia [Kendler etal., 1994].

Categories D1–D5 (intermediate; 700 affecteds, 420affected sib pairs) involved a narrow definition of theschizophrenia spectrum, including disorders whichhave been repeatedly shown to coaggregate in familieswith narrowly defined schizophrenia [Kendler et al.,1993c]. This category added to the narrow definitionschizotypal personality disorder, and all other nonaf-fective psychotic disorders (i.e., schizophreniform dis-order, delusional disorder, atypical psychosis, and goodoutcome SAD).

Categories D1–D8 (broad; 754 affecteds, 505 affectedsib pairs) involved a broad definition of the schizophre-nia spectrum and included all disorders which signifi-cantly aggregated in relatives of schizophrenic pro-bands in the Roscommon Family Study [Kendler et al.,1993b–d], an epidemiologic, controlled family studyconducted in parallel in the west of Ireland. This cat-egory added to the intermediate definition mood incon-gruent and mood congruent psychotic affective illness,and paranoid, avoidant, and schizoid personality dis-order.

Categories D1–D9 (very broad; all psychiatric disor-ders; 961 affecteds) included all other psychiatric dis-orders (e.g., nonpsychotic affective disorders, anxietydisorders, alcoholism, and other nonschizophreniaspectrum personality disorders).

Genotyping

DNA was extracted directly from blood, as well asfrom lymphoblastoid cell lines which were establishedby standard methods. PCR amplification and gel analy-sis of microsatellite markers were performed as de-scribed elsewhere [Straub et al., 1993].

Schizophrenia Locus in 10p15–p11 297

Linkage Analysis

Pairwise lod scores were calculated under the as-sumption of heterogeneity (which we term H-LODs),using the admixture test as implemented in the pro-gram MENDEL [Lange et al., 1988], with modificationsmade by C.J.M. and L. Ploughman. The parametersused have been published [Straub et al., 1995] and wehave kept these parameters constant for all work ex-cept where noted. We used four genetic models: domi-nant (Dom), intermediate heterozygotic on the pen-etrance scale (Pen), additive on the liability scale (Lia),and recessive (Rec). We present only the maximum H-LOD score for each marker. Estimates of the recombi-nation fractions (u) and proportions of families segre-gating the disease locus (a) are available on request.Using the GENEHUNTER program [Kruglyak et al.,1996], multipoint H-LODs and nonparametric linkage(NPL) scores were generated over the entire region(where results are expressed in lod and normal deviate(z) units, respectively). We report only the NPLall sta-tistic, which is calculated based on groupwise S(all)allele sharing among affecteds in sibships. When infor-mation is missing, the NPL test has been found to beoverly conservative [Kong and Cox, 1997; Davis andWeeks, 1997]. However, this inaccuracy will be smallerin multipoint tests when dense maps are employedthan when pairwise NPL scores are calculated. Withmultipoint analysis, information content is about 70%for most of the region reported on here (data notshown). We have not corrected for multiple testing, forreasons given in a previous publication [Straub et al.,1995].

Both the lod and the H-LOD statistics approximatethe chi-square distribution, with one and two degreesof freedom, respectively. Splitting the total sample of265 families into three equal subsets undoubtedlyweakens the accuracy of this approximation, but giventhe relatively large subsets (n 4 87 or 88) involved, wedo not expect this loss to be significant. We have ana-lyzed a number of the variables likely to be important(e.g., family size distribution, number of affecteds, phe-

notypic classifications, and geographical location) and,as would be expected since the families were randomlyassigned to the subsets, we have not found any notabledifferences.

Estimates of the Fraction of Families Linked

Estimates were taken from the multipoint (GENE-HUNTER) H-LOD and NPL output files.

RESULTS AND DISCUSSION

The markers used and genetic map are shown inTable I. The number of alleles, heterozygosity, andpolymorphism information content (PIC) were thoseobserved in the ISHDSF, and the map was constructedusing the genotyping data. Twelve microsatellite mark-ers spanning 32 centimorgans (cM: sex averaged,Kosambi mapping function) were tested. The distancesin our map agree reasonably well with the Marshfieldgenetic map (http://www.marshmed.org/genetics/), al-though our intermarker distance between D10S611–D10S1426 is 0.4 cM and not 4.8 cM. The largest gap inthe sex averaged map is 10 cM, between D10S2325–D10S674. The marker order also agrees with the MITphysical maps (http://www-genome.wi.mit.edu). Esti-mates of the physical distance for the 11 cM betweenD10S674–D10S183 vary considerably. The three esti-mates we made were: 16 megabases (data from theLDB: http://cedar.genetics.soton.ac.uk/public_html/),18.7 megabases (data from the SHGC radiation hybridmap; http://www-shgc.stanford.edu/), and 23 mega-bases (data from the MIT radiation hybrid map).

Table II shows the maximum pairwise lod scores(found at any recombination fraction and proportion offamilies linked) calculated under the assumption ofheterogeneity (which we term H-LODs), using an ad-mixture test [Smith, 1963] as implemented by the pro-gram MENDEL [Lange et al., 1988]. Four diagnosticcategories [Kendler et al., 1996b] were used: D1–D2(narrow), D1–D5 (intermediate), D1–D8 (broad), andD1–D9 (very broad). Under each diagnostic category,we used four genetic models [Straub et al., 1995]: dom-

TABLE I. Chromosome 10p15–p11 Genetic Markers and Map*

Name LocusNumberof alleles HZ PIC

Distances (Kosambi cM)

Female Male Ratio

Sex-averaged

Intermarkerdistance Map position

AFM063xf4 D10S189 6 0.68 0.63 0ATA31G11 D10S1412 9 0.74 0.70 5.8 6.8 0.85 6.3 6.3GGAA8G02 D10S1216 10 0.74 0.70 3.5 3.9 0.90 3.7 10.0GAAT5F06 D10S2325 10 0.86 0.85 3.6 2.0 1.80 2.8 12.8GATA6E06 D10S674 9 0.78 0.75 11.9 4.2 2.83 10.0 20.8GATA70E11 D10S1423 7 0.77 0.74 3.7 3.3 1.12 3.5 24.3MFD248 D10S245 9 0.73 0.69 2.2 0.9 2.44 1.5 25.8GAAT13D02 D10S2443 6 0.64 0.59 2.2 0.9 2.44 1.6 27.4GGAA7H02 D10S1215 25 0.89 0.88 4.1 1.7 2.41 2.9 30.3GATA3G07 D10S611 6 0.66 0.59 1.4 0.7 2.00 1.0 31.3GATA73E11 D10S1426 7 0.75 0.71 0.3 0.4 0.75 0.4 31.7MFD200 D10S183 17 0.87 0.86 0.3 0.4 0.75 0.3 32.0

*The number of alleles, heterozygosity (Hz), and polymorphism information content (PIC) for each marker are derived from ISHDSF genotyping data,as is the map which was constructed and error-checked with the program CRI-MAP [Donis-Keller et al., 1987].

298 Straub et al.

inant (Dom), additive (Pen and Lia), and recessive(Rec). For most markers, under a given diagnostic cat-egory, the H-LOD scores were quite similar for all fourgenetic models. The largest pairwise H-LOD score was1.95 (P 4 0.005) with marker D10S2443, under diag-nostic category D1–D5 and the recessive genetic model.While the results under D1–D5 and D1–D8 were fairlysimilar, expanding the disease definition to D1–D9 orrestricting it to D1–D2 produced substantially smallerscores, with a number of markers failing to yield posi-tive values at any recombination fraction.

Table III shows the pairwise NPLall z scores. As weobserved on chromosomes 5, 6, and 8, compared to thelod score analysis, the nonparametric results are ingeneral considerably weaker: only two markers,D10S674 (P 4 0.062) and D10S2443 (P 4 0.066) (un-der diagnostic category D1–D5), yielded P values lessthan 0.1. When the data are not fully informative, as isthe case in this pairwise analysis, the NPL test hasbeen shown to be overly conservative [Kong and Cox,1997; Davis and Weeks, 1997].

Figure 1 shows the multipoint H-LODs generatedusing the software GENEHUNTER [Kruglyak et al.,1996] for the diagnostic category D1–D5 and all fourgenetic models. D10S189 is placed at map positionzero, and proximal (centromeric) is to the right. For allgenetic models, there is a broad peak extending over

the 11 cM from D10S674–D10S1426, with the recessivemodel strongest and maximizing at 1.91 (P 4 0.006).The information content, as estimated by GENE-HUNTER, is about 70% for most of the region (data notshown).

Figure 2 shows multipoint H-LODs under the reces-sive genetic model for the four diagnostic categories.Concordant with the pairwise results, the support forlinkage is the greatest under D1–D5 (as in Fig. 1). Di-agnostic categories D1–D8 and D1–D9 produced a sec-ond, smaller peak about 27 cM distal to the primarypeak near D10S2443.

Figure 3 shows the multipoint nonparametric link-age (NPLall) results from GENEHUNTER. The largestpeak (maximum 1.88, P 4 0.03) is in the same locationnear D10S2443 as in the parametric analyses, but thepositive evidence is spread distally over a muchbroader region. Based on the parametric multipoint re-sults, a tentative guess of the most likely location of theputative gene is the 11 cM region between D10S674–D10S1426.

Estimates from the multipoint analysis by GENE-HUNTER of the proportion of families linked rangefrom 5–15% (data not shown). However, while we havedeveloped evidence supporting multiple genes forschizophrenia in these Irish families, using linkage re-sults alone, we have not been able to derive much use-

TABLE II. Pairwise H-LOD Scores for Schizophrenia and Chromosome 10p15–11 Markers*

D1–D2 D1–D5 D1–D8 D1–D9

Dom Pen Lia Rec Dom Pen Lia Rec Dom Pen Lia Rec Dom Pen Lia Rec

D10S189 0.00 0.00 0.05 0.26 0.03 0.02 0.00 0.19 0.26 0.12 0.03 0.14 0.06 0.10 0.15 1.22D10S1412 0.07 0.00 0.00 0.00 0.36 0.12 0.17 0.06 0.86 0.45 0.54 0.93 0.38 0.36 0.43 1.27D10S1216 0.00 0.00 0.00 0.00 0.13 0.07 0.16 0.87 0.16 0.10 0.10 0.85 0.00 0.00 0.00 0.23D10S2325 0.05 0.04 0.03 0.01 0.98 0.75 0.97 0.98 0.65 0.47 0.34 0.36 0.19 0.27 0.21 0.08D10S674 0.36 0.40 0.45 0.25 1.08 1.16 1.38 1.30 0.88 0.86 0.88 1.18 0.15 0.27 0.32 0.62D10S1423 0.54 0.42 0.31 0.22 0.81 0.83 1.22 1.46 0.61 0.68 0.68 1.10 0.28 0.23 0.06 0.27D10S245 0.02 0.03 0.06 0.01 0.25 0.36 0.40 0.31 0.02 0.06 0.03 0.04 0.27 0.18 0.03 0.00D10S2443 0.69 0.67 0.57 0.58 1.69 1.56 1.77 1.95 0.85 0.74 0.91 1.33 0.35 0.45 0.55 0.11D10S1215 0.28 0.31 0.46 0.45 0.60 0.63 0.79 1.31 0.29 0.31 0.37 0.97 0.27 0.34 0.35 0.12D10S611 0.00 0.00 0.00 0.03 0.52 0.45 0.52 0.69 0.15 0.09 0.08 0.35 0.20 0.13 0.12 0.26D10S1426 0.67 0.69 0.92 0.65 0.56 0.67 0.99 1.07 0.36 0.45 0.75 1.21 1.14 0.92 0.86 0.38D10S183 0.43 0.34 0.43 0.58 0.94 0.99 1.27 1.65 0.54 0.50 0.59 1.15 1.04 0.92 0.61 0.35

*The program MENDEL [Lange et al., 1988] was used to calculate lod scores under the assumption of locus heterogeneity (H-LODs), which are presentedas the maximum lod scores at any recombination fraction and proportion-linked. Diagnostic categories and genetic models are as described in Materialsand Methods.

TABLE III. Pairwise NPL Scores for Schizophrenia and Chromosome 10p15–11 Markers*

Locus

D1–D2 D1–D5 D1–D8 D1–D9

NPL P value NPL P value NPL P value NPL P value

D10S189 0.23 0.40 0.18 0.42 0.13 0.44 0.13 0.44D10S1412 0.09 0.46 0.58 0.27 0.72 0.23 0.23 0.40D10S1216 0.12 0.44 0.54 0.29 0.49 0.30 −0.33 0.62D10S2325 0.32 0.36 1.04 0.14 0.61 0.26 0.44 0.32D10S674 0.74 0.21 1.50 0.06 1.06 0.14 0.45 0.32D10S1423 0.73 0.21 1.10 0.13 0.79 0.21 0.13 0.44D10S245 0.12 0.44 0.59 0.27 0.16 0.43 0.38 0.35D10S2443 1.01 0.13 1.47 0.07 1.15 0.12 0.71 0.23D10S1215 0.73 0.21 1.16 0.12 0.98 0.16 0.53 0.29D10S611 0.06 0.47 0.48 0.31 0.10 0.45 0.18 0.42D10S1426 0.82 0.18 1.09 0.13 0.89 0.18 1.07 0.14D10S183 0.90 0.16 1.67 0.04 1.25 0.10 1.16 0.12

*The program GENEHUNTER [Kruglyak et al., 1996] was used to calculate pairwise NPLall z scores.

Schizophrenia Locus in 10p15–p11 299

ful information about the nature of their interaction.Since such interaction effects have the potential to con-found the relationship between gene effects and lodscore, we caution against the overinterpretation ofthese GENEHUNTER values of ‘‘fraction-linked.’’ Withthe exception of the regions of interest on 5q, 6p, and8p, our maximum multipoint H-LOD of 1.91 on 10p ishigher than for any other region that we have tested.Although the statistical support for this region is mod-est, when viewed in the context of our genome scan, itprobably deserves attention.

There have not been, to our knowledge, any previouspublications of positive linkage findings for schizophre-nia in this chromosomal region. During preparation ofthis manuscript, in October 1997, we presented some ofour preliminary positive results [Straub et al., 1997a]and learned of positive results in this region [Faraoneand NIMH Genetics Initiative—Millennium Schizo-phrenia Consortium, 1997]. In complete genome scan

of the 48 European-American schizophrenia familiesfrom the NIMH Genetics Initiative, the region contain-ing markers D10S1423 (NPL z 4 3.4, P 4 0.003) andD10S582 (NPL z 4 3.3, P 4 0.0005) yielded the great-est evidence of linkage. The most positive markers inthe two studies are in similar locations, but our pri-mary multipoint peak covers a much more limited re-gion, indicating a target region of less than 20 cM. Inaddition, we have since learned that D. Wildenauer(Univeristy of Bonn) has generated positive resultswith markers D10S582 (mean IBD sharing of 61.5%; P4 0.0058) and D10S1423 (mean IBD sharing of 59%; P4 0.029) in 72 families (59 German and 13 non-Ashkenazi Israeli). Taken together, these three studiessuggest that an additional vulnerability locus forschizophrenia may be located in 10p15–p11.

ACKNOWLEDGMENTS

Data collection was conducted under the supervisionof S. Humphries, M. Healy, and A. Finnerty. Additionalinterviews were conducted by J. Burke, B. Murphy, F.Duke, R. Shinkwin, M. Ni Nuallain, F. McMahon, J.Downing, T. Hebron, B. Hanratty, E. Crowe, M.Doherty, J. Bray, and L. Lowry. This project would nothave been possible without the cooperation of the fami-lies themselves and the staffs of the many psychiatrichospitals and units in Ireland and Northern Ireland,and their efforts are gratefully acknowledged. We alsoacknowledge the assistance of B. Kadambi, B. Worm-ley, H. Sadek, and R. McClelland.

REFERENCESAmerican Psychiatric Association (1987): ‘‘Diagnostic and Statistical

Manual of Mental Disorders.’’ Washington, DC: American PsychiatricAssociation.

Cannon TD, Kaprio J, Lonnqvist J, Huttunen M, Koskenvuo M (1998): Thegenetic epidemiology of schizophrenia in a Finnish twin cohort. ArchGen Psychiatry 55:67–74.

Cloninger CR (1994): Turning point in the design of linkage studies ofschizophrenia. Am J Med Genet 54:83–92.

Davis S, Weeks DE (1997): Comparison of nonparametric statistics for

Fig. 1. Multipoint heterogeneity lod scores based on 12 markers in the10p15–p11 region for the intermediate (D1–D5) disease definition and allfour genetic models. Marker D10S189 is arbitrarily set at map positionzero, and the locations of the labels for the markers are approximate. SeeTable I for locations and intermarker distances. Calculations were per-formed by the GENEHUNTER program [Kruglyak et al., 1996], and areexpressed in lod units. Right is proximal (centromeric).

Fig. 2. Multipoint heterogeneity lod scores under a recessive mode ofinheritance with all four diagnostic categories.

Fig. 3. Multipoint nonparametric linkage (NPLall) scores for the narrow(D1–D2), intermediate (D1–D5), broad (D1–D8), and very broad (D1–D9)diagnostic categories. Calculations were performed by the GENEHUNTERprogram [Kruglyak et al., 1996], and are expressed as normal deviates (zscores).

300 Straub et al.

detection of linkage in nuclear families: single-marker evaluation. AmJ Hum Genet 61:1431–1444.

Donis-Keller H, Green P, Helms C, Cartinhour S, Weiffenbach B, StephensK, Keith TP, Bowden DW, Smith DR, Lander ES, Botstein D, Akots G,Rediker KS, Gravius T, Brown VA, Rising MB, Parker C, Powers JA,Watt DE, Kauffman ER, Bricker A, Phipps P, Muller-Kahle H, FultonTR, Ng S, Schumm JW, Braman JC, Knowlton RG, Barker DF, CrooksSM (1987): A genetic linkage map of the human genome. Cell 51:319–337.

Faraone SV, NIMH Genetics Initiative—Millennium Schizophrenia Con-sortium (1997): A genome scan of the European-American schizophre-nia pedigrees of the NIMH Genetics Initiative. Am J Med Genet 74:557–557.

Gill M, Vallada H, Collier D, Sham P, Holmans P, Murray R, McGuffin P,Nanko S, Owen M, Antonarakis S, Housman D, Kazazian H, NestadtG, Pulver AE, Straub RE, MacLean CJ, Walsh D, Kendler KS, DeLisiL, Polymeropoulos M, Coon H, Byerley W, Lofthouse R, Gershon E(1996): A combined analysis of D22S278 marker alleles in affected sib-pairs: Support for a susceptibility locus for schizophrenia at chromo-some 22q12. Am J Med Genet 67:40–45.

Gottesman II, Shields J (1982): ‘‘Schizophrenia: The Epigenetic Puzzle.’’New York: Cambridge University Press.

Karayiorgou M, Gogos JA (1997): Dissecting the genetic complexity ofschizophrenia. Mol Psychiatry 2:211–223.

Kendler KS, Diehl SR (1993): The genetics of schizophrenia: A current,genetic-epidemiologic perspective. Schizophr Bull 19:261–285.

Kendler KS, Lieberman JA, Walsh D (1989): The Structured Interview forSchizotypy (SIS): A preliminary report. Schizophr Bull 15:559–571.

Kendler KS, McGuire M, Gruenberg AM, O’Hare A, Spellman M, Walsh D(1993a): The Roscommon family study: IV. Affective illness, anxietydisorders, and alcoholism in relatives. Arch Gen Psychiatry 50:952–960.

Kendler KS, McGuire M, Gruenberg AM, O’Hare A, Spellman M, Walsh D(1993b): The Roscommon family study: I. Methods, diagnosis of pro-bands, and risk of schizophrenia in relatives. Arch Gen Psychiatry50:527–540.

Kendler KS, McGuire M, Gruenberg AM, O’Hare A, Spellman M, Walsh D(1993c): The Roscommon family study: III. Schizophrenia-related per-sonality disorders in relatives. Arch Gen Psychiatry 50:781–788.

Kendler KS, McGuire M, Gruenberg AM, Spellman M, O’Hare A, Walsh D(1993d): The Roscommon Family Study: II. The risk of nonschizo-phrenic nonaffective psychoses in relatives. Arch Gen Psychiatry 50:645–652.

Kendler KS, McGuire M, Gruenberg AM, Walsh D (1994): An epidemio-logic, clinical, and family study of simple schizophrenia in CountyRoscommon, Ireland. Am J Psychiatry 151:27–34.

Kendler KS, McGuire M, Gruenberg AM, Walsh D (1995): Examining thevalidity of DSM-III-R schizoaffective disorder and its putative subtypesin the Roscommon family study. Am J Psychiatry 152:755–764.

Kendler KS, MacLean CJ, O’Neill FA, Burke J, Murphy B, Duke F, Shink-win R, Easter SM, Webb BT, Zhang J, Walsh D, Straub RE (1996a):Evidence for a schizophrenia vulnerability locus on chromosome 8p inthe Irish Study of High-Density Schizophrenia Families. Am J Psychia-try 153:1534–1540.

Kendler KS, O’Neill FA, Burke J, Murphy B, Duke F, Straub RE, ShinkwinR, Nuallain MN, MacLean CJ, Walsh D (1996b): Irish Study of High-Density Schizophrenia Families: Field methods and power to detectlinkage. Am J Med Genet 67:179–190.

Kong A, Cox NJ (1997): Allele-sharing models: LOD scores and accuratelinkage tests. Am J Hum Genet 61:1179–1188.

Kruglyak L, Daly MJ, Reeve-Daly MP, Lander ES (1996): Parametric andnonparametric linkage analysis: A unified multipoint approach. Am JHum Genet 58:1347–1363.

Lange K, Weeks D, Boehnke M (1988): Programs for pedigree analysis:MENDEL, FISHER, and dGENE. Genet Epidemiol 5:471–472.

O’Rourke DH, Gottesman II, Suarez BK, Rice J, Reich T (1982): Refutationof the general single-locus model for the etiology of schizophrenia. AmJ Hum Genet 34:630–649.

Prescott CA, Gottesman II (1993): Genetically mediated vulnerability toschizophrenia [review]. Psychiatr Clin North Am 16:245–267.

Pulver AE, Karayiorgou M, Wolyniec PS, Lasseter VK, Kasch L, Nestadt G,Antonarakis S, Housman D, Kazazian HH, Meyers D, Ott J, Lamacz M,Liang K-Y, Hanfelt J, Ullrich G, DeMarchi N, Ramu E, McHugh PR,Adler L, Thomas M, Carpenter WT, Manschreck T, Gordon CT, Kim-berland M (1994): Sequential strategy to identify a susceptibility genefor schizophrenia: Report of potential linkage on chromosome 22q12–q13.1: Part 1. Am J Med Genet 54:36–43.

Pulver AE, Lasseter VK, Kasch L, Wolyniec P, Nestadt G, Blouin J-L,Kimberland M, Babb R, Vourlis S, Chen H, Lalioti M, Morris MA,Karayiorgou M, Ott J, Meyers D, Antonarakis SE, Housman D, Kaza-zian HH (1995): Schizophrenia: A genome scan targets chromosomes3p and 8p as potential sites of susceptibility genes. Am J Med Genet60:252–260.

Schizophrenia Linkage Collaborative Group for Chromosomes 3 and 6 and8 (1996): Additional support for schizophrenia linkage on chromosomes6 and 8: A multicenter study. Am J Med Genet 67:580–594.

Smith CAB (1963): Testing for heterogeneity of recombination fractionvalues in human genetics. Ann Hum Genet 27:175–182.

Spitzer RL, Williams JBW, Gibbon M, First MB (1992): The structuredclinical interview for DSM-III-R (SCID): I: History, rationale, and de-scription. Arch Gen Psychiatry 49:624–629.

Straub RE, Speer MC, Luo Y, Rojas K, Overhauser J, Ott J, Gilliam TC(1993): A microsatellite genetic linkage map of human chromosome 18.Genomics 15:48–56.

Straub RE, MacLean CJ, O’Neill FA, Burke J, Murphy B, Duke F, Shink-win R, Webb BT, Zhang J, Walsh D, Kendler KS (1995): A potentialvulnerability locus for schizophrenia on chromosome 6p24–22: Evi-dence for genetic heterogeneity. Nat Genet 11:287–293.

Straub RE, MacLean CJ, O’Neill A, Walsh D, Kendler KS (1997a): Genomescan for schizophrenia genes: A detailed progress report. Am J MedGenet 74:558.

Straub RE, MacLean CJ, O’Neill FA, Walsh D, Kendler KS (1997b): Sup-port for a possible schizophrenia vulnerability locus in region 5q21–31in Irish families. Mol Psychiatry 2:148–155.

Wang S, Sun C, Walczak CA, Ziegle JS, Kipps BR, Goldin LR, Diehl SR(1995): Evidence for a susceptibility locus for schizophrenia on chromo-some 6pter–6p22. Nat Genet 10:41–46.

Schizophrenia Locus in 10p15–p11 301