Additional file 4 of Exploring autism spectrum disorder and co-occurring trait associations to elucidate multivariate genetic mechanisms and insights

Supplementary Table 1. Data and sample details of ASD and 8 genetically correlated traits (P < 0.05, calculated from LD Score Regression (LDSC)) are presented and applied towards multivariate-GWAS. Data from four excluded traits are additionally shown. Supplementary Table 2. 37 multivariate associations are identified with ASD as a central trait where 17/37, shown with asterisk are previously reported in the GWAS Catalog and in bold, 8 genes are identified as SFARI ASD genes. Supplementary Table 3. 19 gene regions/trait pairings passed coloc (Posterior Prob H4 > 90%, Shown in bold) called on coloc.abf with a window size of ± 50 KB flanking the SNP locus. Supplementary Table 4. (A) Mendelian randomization (MR) results for ASD as outcome and related traits. (B) MR where ASD is the exposure and related traits are the outcome. Supplementary Table 5. MV associated genes are found in systems curated/implicated with gut microbiome and neural systems from GeneCards. Supplementary Table 6. List of 637 Significant SNPs (p < 5e-8), with 315 already reported in the GWAS catalog, identified by MetaPhat multivariate-GWAS using ASD and 8 genetically correlated trait summary statistics. Supplementary Table 7. A) 108 enriched (p < 0.05) Go terms are annotated and (B) 46 pathways on WikiPathway C) KEGG D) Reactome resources e) Tissue from the list of multivariate ASD SNPs found enrichments in neuron and nervous systems related data. Supplementary Table 8. ASD central SNP alleles are mapped to GEMMA genotypes called from 112 (49% females) WGS samples (45 (42% females) ASD probands). Phi coefficients are calculated between allele proportions where Chi-square test is applied to assess statistical importance. Indicated with *. Fisher's exact test is applied when Chi-square assumptions are violated. Supplementary Table 9. eQTL and sQTL related results of the ASD central associations relative to brain and nervous systems from EBI QTL catalog are captured via https://fivex.sph.umich.edu/. Study URLs are listed at the bottom of the table.

补充表1. 孤独症谱系障碍（ASD, Autism Spectrum Disorder）以及经连锁不平衡得分回归（LD Score Regression, LDSC）计算得到遗传相关且P<0.05的8个性状的数据与样本详情已呈现，并将其应用于多变量全基因组关联分析（multivariate-GWAS）。此外还额外展示了4个被排除性状的数据。
补充表2. 以ASD为核心性状，共鉴定出37个多变量关联信号，其中17/37（标注有星号）已在全基因组关联分析目录（GWAS Catalog）中报道，且以粗体标注；另有8个基因被鉴定为SFARI孤独症谱系障碍基因。
补充表3. 共19个基因区域-性状配对通过共定位分析（coloc）筛选，其共定位后验概率H4>90%，以粗体标注；该分析通过coloc.abf工具完成，设定的窗口范围为SNP位点侧翼±50KB。
补充表4. (A) 以ASD为结局、相关性状为暴露的孟德尔随机化（Mendelian randomization, MR）分析结果；(B) 以ASD为暴露、相关性状为结局的孟德尔随机化分析结果。
补充表5. 在GeneCards数据库中经人工整理注释并与肠道微生物组及神经系统相关的系统内，鉴定到了多变量关联基因。
补充表6. 通过MetaPhat工具，利用ASD及8个遗传相关性状的汇总统计数据开展多变量全基因组关联分析，共鉴定得到637个显著SNPs（p<5e-8），其中315个已被收录于全基因组关联分析目录。
补充表7. (A) 共注释得到108个经富集分析筛选（p<0.05）的基因本体（Gene Ontology, GO）术语；(B) 46条维基通路（WikiPathway）、(C) 京都基因与基因组百科全书（KEGG）、(D) Reactome通路数据库以及(E) 组织层面的富集分析结果：基于多变量ASD相关SNPs的分析显示其在神经元及神经系统相关数据中存在显著富集。
补充表8. 将ASD核心SNP等位基因与来自112例（女性占比49%）全基因组测序（Whole Genome Sequencing, WGS）样本的GEMMA基因型进行比对；该样本包含45例（女性占比42%）ASD先证者。计算了等位基因频率间的φ（Phi）系数，并通过卡方检验评估统计学显著性，标注有星号的结果为经该检验得到的显著结果。当卡方检验的假设条件不满足时，则采用费希尔精确检验（Fisher's exact test）。
补充表9. 通过https://fivex.sph.umich.edu/ 工具，从欧洲生物信息研究所（EBI）QTL数据库中提取了与ASD核心关联相关的表达数量性状位点（expression Quantitative Trait Locus, eQTL）及剪接数量性状位点（splicing Quantitative Trait Locus, sQTL）分析结果，这些结果均与大脑及神经系统相关。各研究的URL已列于表格底部。