Additional file 1 of Polygenic risk for triglyceride levels in the presence of a high impact rare variant

Additional file 1: Table S1. Deletion extent details of individuals with a typical pathogenic 22q11.2 deletion included in the main polygenic risk score analyses (nmax=151). Table S2. Comparison of lipid and other clinical and demographic variables between the adult 22q11.2DS cohort included in the main PRS analyses in this study (maximum n=151) and in Blagojevic et al. (2022) [1] (n=267) (n=125 overlapping). Table S3. Polygenic risk scores developed in Privé et al. 2022 [3] (PGS ID: PGP000263) for each lipid trait, retrieved from the PGS catalog [4]. Table S4. Linear regression models testing an interaction between TG-PRS and BMI in adults with a 22q11.2 microdeletion. Table S5. Genes associated with each lipid level that are included in the Dron et al. 2020 [2] NGS panel. This table is adapted from Dron et al. 2020 [2]. Table S6. Within individuals of European ancestry with a 22q11.2 microdeletion, linear regression analyses testing lipid polygenic risk score (PRS) as a predictor of its corresponding lipid level in a univariable model and in a multivariable model accounting for phenotypic, batch, and ancestry variables. Table S7. Measures related to receiver operating characteristic (ROC) curves based on logistic regression models for mild-moderate hypertriglyceridemia (triglyceride level 1.7–10.0 mmol/L; outcome variable) in adults with 22q11.2DS using variables in the “Predictor variables” column. Table S8. Comparison of the area under the curve (AUC) between two receiver operating characteristic (ROC) curves constructed using logistic regression models that predict mild-moderate hypertriglyceridemia status (triglyceride level 1.7–10.0 mmol/L) (outcome variable) in adults with 22q11.2DS. Table S9. Thirty-eight very rare (PopMax FAF<0.2%) LoF or predicted deleterious missense heterozygous variants in 32 individuals with 22q11.2DS prioritized for assessment of clinical relevance with respect to extreme levels of TG, LDLC or HDLC. Table S10. Variant assessments based on ACMG/AMP guidelines for the two variants classified as pathogenic/likely pathogenic. Table S11. Variants (frequency<1.0%) in adults with 22q11.2DS included in the rare variant burden analyses. Table S12. Linear regression analyses testing lipid trait-specific rare carrier status as a predictor for its corresponding lipid level in a univariable model and in a multivariable model accounting for phenotypic, batch, ancestry variables, in adults with 22q11.2DS. Figure S1. Overview of study design. Figure S2. First two principal components of individuals with 22q11.2DS from the TCAG (n=88) and IBBC (n=63) cohorts, and individuals from 1000 Genomes with assigned ancestry. A cut-off of -0.02 on the PC2 axis was used to assign individuals in this study (TCAG or IBBC) as European (<-0.02) or non-European (>0.02). Figure S3. Distribution of polygenic risk scores (PRS) for triglycerides (TG) (n=149), high density lipoprotein cholesterol (HDLC) (n=150), low density lipoprotein cholesterol (LDLC) (n=148), and total cholesterol (TC) (n=151). Figure S4. Distribution of triglyceride levels in individuals with a 22q11.2 deletion (n=149), before and after log transformation. Figure S5. Number of rare variants called per individual between the three site-platform groups, by variant type, for n=151 individuals included in the rare variant burden analyses. The mean ± SD number of variants across all three groups (n=151) is displayed next to each plot title. Variant count means between the three site-platform groups were compared using the Kruskal–Wallis test and pairwise comparisons were made using Wilcoxon signed-rank without adjustment for multiple comparisons. Samples were sequenced to average ± standard deviation (SD) read depths of 29.36 ± 8.42 (IBBC-HiSeq 2500), 37.22 ± 10.90 (IBBC-HiSeq X) and 94.58 ± 23.33 (TCAG-HiSeq X). The read depths reported for the TCAG-HiSeq X samples are the combined read depths of the proband and both parents. Figure S6. Lipid levels of individuals from each site-platform group for n=151 individuals included in the polygenic risk score and rare variant burden analyses. Lipid levels between the three site-platform groups per trait were compared using the Kruskal-Wallis test and pairwise comparisons were made using Wilcoxon signed-rank without adjustment for multiple comparisons. TG, triglyceride; LDLC, low density lipoprotein cholesterol; HDLC, high density lipoprotein cholesterol; TC, total cholesterol. Figure S7. Linear association between the polygenic risk score (common variants) for each lipid trait and its corresponding lipid level. Fitted lines were generated using linear regression performed within sex. TG, triglyceride; LDLC, low density lipoprotein cholesterol; HDLC, high density lipoprotein cholesterol; TC, total cholesterol; PRS, polygenic risk score. Figure S8. Receiver operating characteristic (ROC) curves of logistic regression models predicting mild-moderate hypertriglyceridemia in adults with 22q11.2DS. Each plot contains two ROC curves. One contains only one or two clinical variable(s) as predictors: BMI (A), sex (B) or both (C). The other curve includes the TG-PRS with the respective clinical variable(s). The area under the curve (AUC) and 95% confidence intervals (95% CI) for each curve are shown in the figure key on the bottom right of each plot. Figure S9. The proportion of adult individuals with 22q11.2DS and (A) mild-moderate hypertriglyceridemia (HTG) and (B) the triglyceride (TG) level (mmol/L) for individuals in each decile of triglyceride polygenic risk score (TG-PRS) (n=14–15 per decile; n=149 total). The dashed line in B indicates the lower-bound cut-off for mild-moderate HTG (1.7 mmol/L).

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.