Abstract P212: Combined Analysis of Longitudinal Cohorts With Case-only Sample Sets for Detecting Genetic Effects on Venous Thromboembolism in Topmed
Bibliographic record
Abstract
Background: The Trans-Omics for Precision Medicine (TOPMed) program is performing whole-genome sequencing across multiple studies, including longitudinal cohorts, case-control, and case-only sample sets. Detecting the effects of low frequency variants requires large sample sizes, which can only be achieved by combining data across diverse study designs, including matching of case-only sample sets with controls from other studies. Here we present a strategy for combined analysis of venous thromboembolism (VTE) case status in 3 longitudinal cohorts and 3 case-only sample sets, in the context of whole-genome association studies. Methods: For each cohort study, we sampled ‘pure’ controls (without replacement) from the risk set of each incident case, within strata defined by sex, ancestry and birth-cohort. Pure controls had no event throughout their period of observation. A case’s risk set was defined as controls with no prior VTE history and under observation through an age at least as old as the case. For case-only sample sets, controls for each case were sampled from a cohort study, using the same risk set definition. Because of limited overlap in birth years between the cohort studies and the case-only sample sets, this matching was done within strata defined only by sex and ancestry group. Mixed model logistic regression will be used to account for relatedness as a random effect. Although conditional logistic regression is not practical for whole-genome association studies, case-control matching is implicitly recognized by a fixed effect for age-at-event for each matched set (1 case and >=1 matched controls). Additional fixed effects will include sample set and sex. We will also adjust for variations in case-control ratio among the matched sets. Results: In two cohort studies, we matched 1,231 cases to 4,820 controls (overall ratio = 1:3.9); only 500 controls and 5 cases could not be matched. For the case-only sample sets we matched 2,141 cases to 2,780 controls from one cohort study (overall ratio = 1:1.3); zero controls and only 14 cases could not be matched. Conclusions: We were successfully able to match nearly all cases to controls, and more than 90% of controls were also matched. By matching controls to cases based on age at event, we can account for the different risk of VTE by age clusters. Although this strategy does not provide an asymptotically unbiased estimated of the hazard ratio, compared to classical risk set sampling, it uses a large portion of the available data, it provides odds ratio estimates yielding the correct sign of association, and it reduces the potential influence of resampling subjects with rare variants. This strategy also enables combined analysis of multiple studies with different designs.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.046 | 0.083 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.008 |
| Bibliometrics | 0.004 | 0.004 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.004 | 0.001 |
| Open science | 0.003 | 0.004 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.009 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".