Multipoint LOD Score Research Articles

Calculation of multipoint likelihoods using flanking marker data: a simulation study

The calculation of multipoint likelihoods is computationally challenging, with the exact calculation of multipoint probabilities only possible on small pedigrees with many markers or large pedigrees with few markers. This paper explores the utility of calculating multipoint likelihoods using data on markers flanking a hypothesized position of the trait locus. The calculation of such likelihoods is often feasible, even on large pedigrees with missing data and complex structures. Performance characteristics of the flanking marker procedure are assessed through the calculation of multipoint heterogeneity LOD scores on data simulated for Genetic Analysis Workshop 14 (GAW14). Analysis is restricted to data on the Aipotu population on chromosomes 1, 3, and 4, where chromosomes 1 and 3 are known to contain disease loci. The flanking marker procedure performs well, even when missing data and genotyping errors are introduced.

Genetic Heterogeneity of Autosomal Dominant Hypercholesterolemia in Mexico

Familial hypercholesterolemia (FH) and familial defective apolipoprotein B-100 (FDB) are relatively common lipid disorders caused by mutations of the low-density lipoprotein receptor (LDLR) and apolipoprotein B (apoB) genes, respectively. A third locus on chromosome 1p34.1-p32 was recently linked to FH and the responsible gene has been identified [protein convertase subtilisin/kexin type 9 (PCSK9)]. We assessed the contribution of the LDLR, apoB, and PCSK9 genes as cause of FH in Mexico. Forty six unrelated probands, as well as 68 affected and 60 healthy relatives, were included. All index cases were diagnosed as having heterozygous autosomal dominant FH. Seventeen of the 46 index cases had LDLR gene mutations, four of which were novel (Fs92ter108, C268R, Q718X, and Fs736ter743); and only one patient had an apoB mutation (R3500Q). We sequenced the PCSK9 gene in the remainder of the 28 probands with no identified LDLR or APOB gene defects; however, no PCSK9 mutations were found, including one large kindred with positive linkage to the 1p34.1-32 locus (multipoint LOD score of 3.3) and two small pedigrees. Linkage was excluded from these three loci in at least four kindreds suggesting that other yet uncharacterized genes are involved. Our results underline substantial genetic heterogeneity for FH in the Mexican population.

Two Candidate Genes for Low Platelet Count Identified by Genome-Wide Linkage Analysis: Glycoprotein IX and Thrombopoietin.

Although the 'normal' range of platelet count in a healthy person is quite broad (150 to 400 ×109/L), an individual's platelet count is maintained within a much narrower range, constantly balanced between thrombopoiesis and senescence. Platelet counts within the normal range have been shown to be largely determined by genetic factors with heritability estimates from twin studies ranging from 0.57 to 0.80. A genome-wide linkage analysis was carried out in a large Asian Indian kindred with β thalassemia to search for genes influencing variation in normal platelet count. Significant linkage is shown at two locations on chromosome 3q with age, sex and β thalassemia adjusted platelet counts. Linkage analysis showed one marker (D3S1309) on chromosome 3q with a lod score of 3.26 and another marker (D3S1282) approximately 30 cM centromeric, with a lod score of 2.52. Multipoint linkage analysis across 90 cM of chromosome 3q identified two peaks across the region with maximum multipoint lod scores of 3.52 and 4.11 under markers D3S1309 and D3S1282, respectively. Two strong candidate genes for platelet variation were identified in the linked region; thrombopoietin (THPO) and glycoprotein IX (GPIX). Resequencing of four individuals revealed five single nucleotide polymorphisms (SNPs) in THPO and one mutation in the transmembrane region of GPIX. Analysis of variance showed that the GPIX mutation and one THPO SNP accounted for 6% and 4% of the variation in platelet count in the kindred, respectively. The THPO SNP lies in the 3′ untranslated region (3′ UTR) of the gene and has not been previously reported.Although no function could be attributed to the THPO SNP it could potentially influence post-transcriptional control of gene expression. The G to A transition in nucleotide 653 (Ref seq NM_000174) resulted in an Ala 156 (GCC) to Thr (ACC) replacement in the GPIX protein. The GPIX mutation was recently identified in a Chinese patient with Bernard-Soulier syndrome (BSS), a rare bleeding disorder characterized by severe thrombocytopenia and giant platelets. The Chinese patient was homozygous for the mutation; however, heterozygous relatives did not show the characteristic BSS symptoms, leading the authors to conclude that the mutation acted under a recessive model of inheritance. Transient transfection studies by the same group confirmed that the mutation prevents GP Ib/IX complex insertion in the cytoplasmic membrane of platelets, and suppresses GPIbα and GPIX expression, but does not cause its intracellular degradation. One copy of the GPIX mutation was found in 300 European individuals with platelet counts within the normal clinical range.Conclusion: Although the genetic determinants of many thrombocythemias and congenital thrombocytopenias have been identified, very little is known about the genetic determinants of normal platelet count variation. The results suggest that two QTLs on chromosome 3q influence platelet count variation in the Asian Indian kindred, with the GPIX transmembrane mutation and the 3′ UTR SNP in THPO, being strong candidates.

Genome-wide linkage of febrile seizures and epilepsy to the FEB4 locus at 5q14.3-q23.1 and no MASS1 mutation

Febrile seizures (FS) represent the most common seizure disorder in childhood and contribution of a genetic predisposition has been clearly proven. In some families FS is associated with a wide variety of afebrile seizures. Generalized epilepsy with febrile seizures plus (GEFS+) is a familial epilepsy syndrome with a spectrum of phenotypes including FS, atypical febrile seizures (FS+) and afebrile generalized and partial seizures. Mutations in the genes SCN1B, SCN1A and GABRG2 were identified in GEFS+ families. GEFS+ is genetically heterogeneous and mutations in these three genes were detected in only a minority of the families. We performed a 10 cM density genome-wide scan in a multigenerational family with febrile seizures and epilepsy and obtained a maximal multipoint LOD score of 3.12 with markers on chromosome 5q14.3-q23.1. Fine mapping and segregation analysis defined a genetic interval of approximately 33 cM between D5S2103 and D5S1975. This candidate region overlapped with a previously reported locus for febrile seizures (FEB4) in the Japanese population, in which MASS1 was proposed as disease gene. Mutation analysis of the exons and exon-intron boundaries of MASS1 in our family did not reveal a disease causing mutation. Our linkage data confirm for the first time that a locus on chromosome 5q14-q23 plays a role in idiopathic epilepsies. However, our mutation data is negative and do not support a role for MASS1 suggesting that another gene within or near the FEB4 locus might exist.

Localization of a novel locus for alopecia with mental retardation syndrome to chromosome 3q26.33–q27.3

Alopecia with mental retardation syndrome is a rare autosomal recessive disorder characterized clinically by total or partial alopecia and mental retardation. In an effort to understand the molecular bases of this form of alopecia syndrome, large Pakistani consanguineous kindred with multiple affected individuals has been ascertained from a remote region in Pakistan. Genome wide scan mapped the disease locus on chromosome 3q26.33-q27.3. A maximum two-point LOD score of 3.05 (theta = 0.0) was obtained at marker D3S3583. Maximum multipoint LOD score exceeding 5.0, obtained with several markers, supported the linkage. Recombination events observed in affected individuals localized the disease locus between markers D3S1232 and D3S2436, spanning 11.49-cM region on chromosome 3q26.33-q27.3. Sequence analysis of a candidate gene ETS variant gene 5 from DNA samples of two affected individuals of the family revealed no mutation.

Two candidate genes for low platelet count identified in an Asian Indian kindred by genome-wide linkage analysis: glycoprotein IX and thrombopoietin

A genome-wide linkage analysis of platelet count was carried out in a large Asian Indian kindred. Linkage analysis showed one marker (D3S1309) on chromosome 3q with a lod score of 3.26 and another (D3S1282) approximately 30 cM centromeric, with a lod score of 2.52. Multipoint analysis of chromosome 3q identified two peaks with maximum multipoint lod scores of 3.52 and 4.11 under markers D3S1309 and D3S1282, respectively. Two strong candidate genes for platelet variation were identified in the linked region; thrombopoietin (THPO) and glycoprotein IX (GPIX). Resequencing of four individuals revealed five single-nucleotide polymorphisms (SNPs) in THPO and one mutation in the transmembrane region of GPIX. Analysis of variance showed that the GPIX mutation and one THPO SNP accounted for 6 and 4% of the variation in platelet count, respectively. The THPO SNP lies in the 3' untranslated region of the gene and has not been previously reported. The G to A transition at nucleotide 653 resulted in an Ala 156 (GCC) to Thr (ACC) replacement in the GPIX protein. The GPIX mutation was recently identified in a Chinese patient with Bernard-Soulier syndrome (BSS), a rare recessive bleeding disorder characterized by thrombocytopenia and giant platelets. One copy of the GPIX mutation was found in 300 European individuals with platelet counts within the normal range. The results suggest that two QTLs on chromosome 3q influence platelet count variation in the Asian Indian kindred, with the GPIX transmembrane mutation and the 3' UTR SNP in THPO being strong candidates.

A genome-wide scan for attention-deficit/hyperactivity disorder in 155 German sib-pairs

Three groups have previously performed genome scans in attention-deficit/hyperactivity disorder (ADHD); linkage to chromosome 5p13 was detected in all of the respective studies. In the current study, we performed a whole-genome scan with 102 German families with two or more offspring who currently fulfilled the diagnostic criteria for ADHD. Including subsequent fine mapping on chromosome 5p, a total of 523 markers were genotyped. The highest nonparametric multipoint LOD score of 2.59 (empirical genome-wide significance 0.1) was obtained for chromosome 5p at 17 cM (according to the Marshfield map). Subsequent analyses revealed (a) a higher LOD score of 3.37 at 39 cM for a quantitative severity score based on symptoms of inattention than for hyperactivity/impulsivity (LOD score of 1.11 at 59 cM), and (b) an HLOD of 4.75 (empirical genome-wide significance 0.001) based on a parametric model assuming dominant inheritance. The locus of the solute carrier 6A3 (SLC6A3; dopamine transporter 1; DAT1) localizes to 5p15.33; the gene has repeatedly been implicated in the etiology of ADHD. However, in our sample the DAT1 VNTR did not show association with ADHD. We additionally identified nominal evidence for linkage to chromosomes 6q, 7p, 9q, 11 q, 12q and 17p, which had also been identified in previous scans. Despite differences in ethnicity, ascertainment and phenotyping schemes, linkage results in ADHD appear remarkably consistent.

Identification of a locus for nongoitrous congenital hypothyroidism on chromosome 15q25.3-26.1

Permanent congenital hypothyroidism is the most prevalent inborn endocrine disorder, and principally due to developmental defects leading to absent, ectopic or hypoplastic thyroid gland. Although commonly regarded as sporadic disease, nonsyndromic thyroid hypoplasia has, in rare cases, been attributed to inherited defects in PAX8 and the TSHR gene. The shared clinical picture caused by these defects is a variable degree of thyrotropin resistance (RTSH [MIM 275200]), accompanied in its severe form by thyroid gland hypoplasia. We recently identified six extended kindreds with autosomal dominant RTSH, only one of which was linked to a mutation in the PAX8 candidate gene. Genome wide scans conducted in two of the remaining five families revealed independently significant linkage to chromosome 15q25.3-26.1, with maximum multipoint LOD scores of 8.51 and 4.31. Linkage to this novel locus was replicated (P<0.01) in each of the three remaining kindreds. Fine mapping of key recombinants in the largest family localized the causative gene within a 3 cM/2.9 Mb interval. Thus, we report the first locus for congenital nongoitrous hypothyroidism identified by a genome wide screening approach.

A High-Density SNP Genomewide Linkage Scan for Chronic Lymphocytic Leukemia–Susceptibility Loci

Chronic lymphocytic leukemia (CLL) and other B-cell lymphoproliferative disorders (LPDs) show clear evidence of familial aggregation, but the inherited basis is largely unknown. To identify a susceptibility gene for CLL, we conducted a genomewide linkage analysis of 115 pedigrees, using a high-density single-nucleotide polymorphism (SNP) array containing 11,560 markers. Multipoint linkage analyses were undertaken using both nonparametric (model-free) and parametric (model-based) methods. Our results confirm that the presence of high linkage disequilibrium (LD) between SNP markers can lead to inflated nonparametric linkage (NPL) and LOD scores. After the removal of high-LD SNPs, we obtained a maximum NPL of 3.14 (P=.0008) on chromosome 11p11. The same genomic position also yielded the highest multipoint heterogeneity LOD (HLOD) score under both dominant (HLOD 1.95) and recessive (HLOD 2.78) models. In addition, four other chromosomal positions (5q22-23, 6p22, 10q25, and 14q32) displayed HLOD scores >1.15 (which corresponds to a nominal P value <.01). None of the regions coincided with areas of common chromosomal abnormalities frequently observed for CLL. These findings strengthen the argument for an inherited predisposition to CLL and related B-cell LPDs.

Linkage exclusion analysis of two important chromosomal regions for height

Adult height (stature), as an important parameter of human physical development, has been studied in many populations. Recently, we reported a whole genome scan of height on a sample of 630 Caucasian subjects from 53 human pedigrees. Two chromosome regions, 6q24–25 and 7q31.3–36, achieved low linkage signals (multipoint LOD score ⩽0.5), but gained significant results in the linkage studies of height by other groups. In addition, the region 6q24–25 harbors the ER-α gene, an important candidate gene for linear growth. To resolve the controversies over these two regions for height, linkage exclusion analyses were performed in an extended sample of 79 pedigrees with 1816 subjects, which include the 53 pedigrees containing 630 subjects for our previous whole genome study and additional 128 new subjects, and 26 new pedigrees containing 1058 subjects. The two regions, 6q24–25 and 7q31.3–36, were excluded at a relative effect size of 10% or greater ( p value < 0.0005) and 5% or greater ( p value < 0.0018), respectively. Our results suggest that the two regions may not contribute substantially to height variation in our Caucasian population.

Genome-wide Scan for Myopia in the Old Order Amish

To identify myopia susceptibility genes influencing common myopia in 34 Old Order Amish families, a genetically well-defined founder population. A prospective study of families with myopia consisting of a minimum of two individuals affected with myopia. Extended families consisting of at least two siblings affected with myopia were ascertained. A genome-wide linkage scan using 387 markers was conducted by the Center for Inherited Disease Research (CIDR). Linkage analyses were conducted with parametric (autosomal dominant, fixed penetrance model) and nonparametric methods. Model-free linkage analysis was also performed maximizing over penetrance and over dominance (that is, fitting a wide range of both dominant and recessive models). Under the fixed penetrance model, the maximum two-point heterogeneity LOD score (HLOD) was 1.59 at D20S451 and the maximum multipoint HLOD was 1.92 at D6S1021. The nonparametric maximum multipoint (NPL) at D3S2427 had a P-value of .0005. Under the model-free analysis, multipoint heterogeneity LOD scores of 2.03 were observed on both chromosomes 8 (under a recessive model between D8S1130 and D8S1106) and X (under a recessive model between DXS6800 and DXS6789). Reanalyses of chromosomes 3, 6, 8, 20, and X using the best penetrance models resulted in maximum multipoint HLODs of 1.84 at D3S3053; 1.84 at D3S2427; 2.04 at D8S1130; and 2.34 at DXS6800. The locus on chromosome 8p23 independently confirms a report by Hammond and associates mapping a myopia quantitative trait loci (QTL) to this region.

A gene locus for progressive familial heart block type II (PFHBII) maps to chromosome 1q32.2-q32.3

Cardiac conduction defects that are associated with dilated cardiomyopathy (DCM) are generally considered to be sporadic clinical entities, although familial forms of disorders with these clinical features have been identified in a number of families in different countries. An autosomal dominant cardiac disorder characterised by conduction abnormalities and DCM, termed progressive familial heart block type II (PFHBII) (OMIM 140400), has been described in a South African Caucasian family of Northern European descent. Known candidate loci for isolated conduction disorders, isolated DCM and conduction disorders complicated by DCM were excluded from disease causation in this family by linkage analysis, with the exception of the DCM-associated (CMD1D) locus on chromosome 1q32, where a maximum multipoint lod score of 3.7 in the interval between D1S3753 and D1S414, was generated. This region encompassed the troponin T gene (TNNT2), however, genetic fine mapping and haplotype analysis excluded TNNT2 as cause of PFHBII and placed the disease-causative gene within a 3.9 cM (2.85 Mb) interval, flanked by D1S70 and D1S505. Analysis of KCNH1, KIAA0205, LAMB3 and PPP2R5A, which map within the critical interval, indicated that the PFHBII-causative mutation does not lie within the coding regions or splice junctions of these plausible candidate genes. The data indicate the existence of a novel locus involved in the pathogenesis of cardiac conduction abnormalities and DCM.

A Novel Quantitative Trait Locus on Chromosome 1 with Pleiotropic Effects on HDL-Cholesterol and LDL Particle Size in Hypertensive Sibships

High-density lipoprotein (HDL)-cholesterol, triglycerides, and LDL particle size are correlated lipid traits. Abnormal levels of these traits are frequent in hypertensive individuals and contribute to increased risk of coronary heart disease (CHD). We performed univariate and bivariate linkage analyses to identify genomic regions that influence levels of these traits and exert pleiotropic effects on the traits in hypertensive sibships. Subjects included 691 non-Hispanic white individuals (mean age 63.1+/-8.5 years, 57% women, 78% hypertensive) ascertained through sibships with two or more individuals diagnosed with hypertension before age 60 years. The LDL particle size was measured by polyacrylamide gel electrophoresis and triglycerides were log-transformed to reduce skewness. Genotypes were measured at 366 microsatellite marker loci distributed across the 22 autosomes. Univariate and bivariate linkage analyses were performed using a variance components approach. Significant (P < .001) genetic correlations were confirmed for all pairwise combinations of the traits. Univariate linkage analyses demonstrated evidence of linkage (defined as multipoint LOD scores > or =1.3) for HDL-cholesterol on chromosomes 1p, 3p, 9q, and 18q; for log triglycerides on chromosome 10q; and for LDL particle size on chromosomes 2p and 8p. Pairwise bivariate linkage analyses of the three traits revealed a region with pleiotropic effects on HDL-cholesterol and LDL particle size on chromosome 1p (LOD score 4.48). These findings indicate the presence of a quantitative trait locus on chromosome 1 that has pleiotropic effects on HDL-cholesterol and LDL particle size and may therefore influence CHD susceptibility in hypertensive sibships.

Genomic search for prostate cancer predisposition loci in Utah pedigrees

We report a genome linkage scan in extended Utah pedigrees, utilizing a pedigree-splitting approach to reduce intra-familial heterogeneity. Fifty-nine pedigrees with at least four Prostate cancer (PrCa) cases and no more than two meioses separating PrCa cases were analyzed using the CIDR genomic search STRP marker set. Parametric linkage analyses using dominant and recessive models were performed on four datasets resulting from a pedigree splitting algorithm. In addition, age at diagnosis subset analyses were performed. Four regions of interest (LODs>1.9) were identified on chromosomes 1p, 3q, 5q, and 22q. The linkage peaks on 1p, 3q, and 22q have been previously implicated for PrCa, though not significantly. The 1p region was supported by a single large Utah pedigree with a multipoint LOD score of 3.1. An additional 10 regions gave LOD scores>1.22 (nominal linkage evidence), including moderate evidence supporting the HPC20 region with a recessive model. Our genome-wide search in the informative, extended Utah pedigrees continues to illustrate an ability to identify and replicate linkage peaks, and supports four regions of interest for PrCa predisposition genes.

A genome-wide screen and linkage mapping for a large pedigree with episodic ataxia

Episodic ataxias are ion channel disorders characterized by attacks of incoordination. The authors performed a genome-wide screen in a large pedigree segregating a novel episodic ataxia and found significant linkage on 1q42 with a multipoint lod score of 3.65. Haplotype analysis and fine mapping yielded a peak 2-point lod score of 4.14 and indicated a 4-cM region on 1q42 that is likely to harbor an episodic ataxia gene.

Haplotype and linkage analysis in Chinese hereditary mixed polyposis syndrome

To investigate if hereditary mixed polyposis syndrome (HMPS) locus of two Singapore Chinese HMPS families is identical with the Ashkenazi families. Genomic DNA was extracted, multiplex polymerase chain reaction (PCR) was used to amplify 4 microsatellite markers D15S1010, D15S1007, ACTC and D15S118 in 31 individuals from two families. The HMPS locus cosecretion of the markers on 15q13 over a region of 2.8 cM was confirmed by Haplotype and linkage analysis. The disease of haplotype identified in one pedigree was not co-segregated with an affected individual while no definitive disease haplotype could be assigned for the second pedigree. The maximum two-point and multi-point LOD scores at ACTC for the two Chinese families are 0.20 (theta = 0.3) and -5.0 respectively. Haplotype and linkage analysis indicate that the Ashkenazi haplotypes is not associated with HMPS in Singapore Chinese families, which suggests genetic heterogeneity.

Gene locus ambiguity in posterior urethral valves/prune-belly syndrome

Lower urinary tract obstruction by posterior urethral valves (PUV) is an important cause of congenital renal failure in male infants. Though population-based studies point to a role of genetic factors in the etiology of PUV, no clear evidence for a specific gene defect for PUV has been observed so far. Here we present a consanguineous family with four male descendants affected by PUV and a healthy girl, suggestive of autosomal recessive inheritance. One boy presented with prune-belly syndrome (PBS) in addition to PUV. Using a DNA chip-based genome-wide linkage analysis, we identified a region of homozygosity for the affected boys in an interval of 35 cM on chromosome 1q41-44 with a maximum multipoint LOD score of Z(max) = 3.134 at theta = 0 for single nucleotide polymorphisms (SNPs) rs158724-rs720163. By applying a second genetic model based on the assumption of a male-limited phenotype and the girl being carrier of the genetic defect without expressing the phenotype, a second alternative locus of 9 cM on chromosome 11p11 was identified with a LOD score of Z(max) = 3.61 at theta = 0. Equal significance for both loci with a LOD score of Z(max) = 3.01 at theta = 0 was obtained after the affection status of the female descendant was set "unknown". We suppose that most probably, only one of the two identified loci harbours the disease-causing gene. As the interpretation of the girl's status remains uncertain, we are not able to exclude one of the two loci. Analyses of additional informative families will be important to exclude one of the two loci and to restrict the critical interval.

Linkage and candidate gene analysis of 14q22-24 in bipolar disorder: support for GCHI as a novel susceptibility gene

Using a collection of Irish sib-pair nuclear families, we previously obtained modest evidence of linkage implicating 14q22-24 in bipolar disorder (BPD). To follow-up on this preliminary finding, an extended linkage analysis was performed which employed thirteen microsatellite markers, spanning a total distance of 85 cM on 14q. Effectively, P-values <0.05 were observed for a region extending over 41.88 cM, with the marker D14S281 displaying a peak multipoint non-parametric lod (NPL) score of 2.72 and an associated P-value of 0.003. Support for this finding was also obtained from flanking markers indicating excess allele sharing at 14q22-24 in Irish bipolar sib-pairs. A web-based candidate gene search of 14q22-24 resulted in the selection of GTP cyclohydrolase I (GCHI), located 200 kb 3' of D14S281, as the best plausible candidate gene for involvement in BPD. GCHI is the rate-limiting enzyme in the biosynthesis of tetrahydrobiopterin (BH(4)), a natural cofactor for tyrosine and tryptophan hydroxylases. These enzymes play an essential role in the biosynthesis of various hormones and neurotransmitters such as dopamine, noradrenaline, adrenaline, and serotonin. Numerous studies have also suggested that the clinical symptoms of depression might be related to a deficiency of BH(4). An association study between BPD and a novel single nucleotide polymorphism (SNP) in GCHI (G to A at position -959 bp, upstream of the ATG codon), is also presented here. This study revealed that the variant A allele is preferentially transmitted to BPI probands (chi(2) = 4.54, P = 0.033) suggesting that variants within GCHI may contribute to BPD in the Irish population.

Localisation of merosin-positive congenital muscular dystrophy to chromosome 4p16.3

The congenital muscular dystrophies (CMD) are a heterogeneous group of autosomal recessive disorders, which present within the first 6 months of life with hypotonia, muscle weakness and contractures, associated with dystrophic changes on skeletal muscle biopsy. We have previously reported a large consanguineous family segregating merosin-positive congenital muscular dystrophy, in which involvement of known CMD loci was excluded. A genome-wide linkage search of the family conducted using microsatellite markers spaced at 10-Mb intervals failed to identify a disease locus. A second scan using a high-density SNP array, however, permitted a novel CMD locus on 4p16.3 to be identified (multipoint LOD score 3.4). Four additional consanguineous CMD families with a similar phenotype were evaluated for linkage to a 4.14-Mb interval on 4p16.3; however, none showed any evidence of linkage to the region. Our findings further illustrate the utility of highly informative SNP arrays compared with standard panels of microsatellite markers for the mapping of recessive disease loci.

Evidence for linkage on 21q and 7q in a subset of autism characterized by developmental regression

Autism is a pervasive developmental disorder with a strong genetic component. While candidate regions of the genome have been identified, location of genes conferring susceptibility to autism has been hindered by the heterogeneity within this clinically defined disorder, and the likely contribution of many genes of weak effect. Subsetting samples on the basis of distinct, nondiagnostic clinical features has been recommended to decrease sample heterogeneity. In this study, linkage analysis was performed on a subset of families in the database of the Autism Genetic Resource Exchange (AGRE). This set of autism-affected relative pairs (n=34) was also concordant for a history of developmental regression as measured by the Autism Diagnostic Interview-Revised (ADI-R). In this sample, a maximum multipoint LOD score of 3.4 under the dominant mode of inheritance and an NPL score of 3.0 (P=1.3 x 10(-3)) were observed on chromosome 21 near D21S1437. On chromosome 7 near D7S483 a LOD score of 2.0 under the dominant mode of inheritance and an NPL score of 3.7 (P=7.9 x 10(-5)) were observed. Genetic elements in these regions of 21q and 7q are likely to confer susceptibility to autism or modify the disease presentation in a subgroup of children characterized by a history of developmental regression.