Online Supplemental Figure & Table Legends
Online supplemental figure 1: SLC6A4 locus, marker set & gene-Linked Polymorphic
Region (LPR)
The gene is transcribed on the (-) strand. 3 SNPS in the marker set (left) lie 5’ to the LPR indel,
between Bleomycin hydrolase and SLC6A4. The most 3’ SNP is downstream of SLC6A4.
Synonymous and nonsynonymous amino acid changes are shown in parenthesis after SNP rs
numbers. The upper left insert (LPR detail) shows the gene-linked polymorphic region (LPR),
the location of rs25531 within the LPR (we have referred to the molecular haplotype at this locus
as LPR-rs25531), and a potential 381-bp deletion between the LPR and transcription start site.
Online supplemental figure 2: SLC6A4 gene-linked polymorphic region (LPR) MspI Digest
Map and 2-step protocol for genotyping rs25531
Methods: Phase-certain haplotyping of the LPR variants La, Lg, Sa and the rare Sg allele was
performed by a 2-step protocol: I. amplification of the LPR for determination of S or L allele;
and, II. digestion of this amplicon with MspI restriction endonuclease. A constant MspI digest
site present within the 419-bp amplicon served as internal control for digestion or partial
digestion. Step II results (rs25531) were obtained only after unambiguous genotyping in step I.
Separation of the digestion products by capillary electrophoresis and analysis (Gene Marker
v1.5) allowed heuristic assemblage of genotypes generated by steps I and II. This resulted in the
following restriction fragment configurations (bp): SS (281), SLa (281/325), SLg (281/151),
LaLa (325), LaLg (325/151), and LgLg (151). Only MspI activity at the variable MspI site
resulted in a 151-bp fragment. A rare Sg FAM-labeled 151-bp fragment obtaining only from the
S allele was detected and confirmed by sequencing. All genotypes were then assessed for
1
discordance and Mendelization-type errors. Step I vs. step II genotype completion rates: 99.6 vs.
96.5%, respectively.
Online supplemental figure 3: Pairwise linkage disequilibrium (LD) relationships (D´)
between the SLC6A4 markers in the OCGS sample and 2 SLC6A4 haplotype blocks
(Haploview v4.1: http://www.broad.mit.edu/node/443)
We observed 2 multi-marker haplotype blocks, consistent with the LD structure observed in the
HapMap dataset (see online supplemental figure 4).
Online supplemental figure 4: Pairwise linkage disequilibrium (LD) relationships (D´)
between SLC6A4–locus markers in the HapMap dataset (phase III)
For reference, the SNP marker set used to genotype the OCGS sample is indicated by
superimposed numbers above the HapMap marker set (see markers 1-11 in online supplemental
figure 3).
Online supplemental table 1: FBAT analysis of LPR & LPR-rs25531 molecular haplotype
1
Case-control data assembled from Hu et al (1); column 2 shows pooled data from Finnish whites and two groups of US whites,
wheras column 4 shows the US whites only.
2
Conditional logistic analysis run in STATA (http://www-gene.cimr.cam.ac.uk/clayton/software/stata) showing genotype relative risk
(GRR) estimates (LR chi-squared test=1.9 on 4 df, p=0.76); insufficiently frequent genotypes vs. the reference genotype generated
a pooled GRR (GRR=1.1, Z=0.23, P=0.82 and 95% CI=0.54-2.17). Calculations are based on 687 genotyped affected offspring and
320 complete trios.
3
LA/LA genotype vs. all other genotypes, i.e., as initially reported by Hu et al1. Other entries in column are single genotype
comparisons vs. reference LA/LA that meet STATA minimum thresholds.
**FBAT family number minimum threshold not achieved; ***STATA minimum threshold not achieved.
Online supplemental table 2: Non-transmitting pedigrees of the rare 425V variant, OCD
phenotypes, and neuropsychiatric comorbidities
1
Red highlight indicates 425V carrier status; yellow highlight indicates OCD status; and orange highlight indicates OCPD status.
Fractions indicate number of DSM-IV diagnostic criteria met.
3
Parenthesis indicates probable diagnosis.
4
Neo measure abbreviations: N=neuroticism, LC=low conscientiousness, E=extroversion, O=openness, A=agreeableness.
2
2
3