AS564 - Trans-Omics for Precision Medicine (TOPMed) - Sample Processing

Investigator Names and Contact Information

Charles Kooperberg (clk@fredhutch.org)

Alex Reiner

Introduction/Intent

For more information on the NHLBI TOPMed program visit: www.nhlbi.nih.gov/research/resources/nhlbi-precision-medicine-initiative/topmed​​

WHI investigators should be encouraged to propose papers using this data, and possibly use those proposals and the TOPMed data as basis to try to acquire analysis grants.

Investigators interested in submitting paper proposals using TOPMed data must submit them for review to the TOPMed P&P rather than to the WHI P&P. Step one of this process is to become a TOPMed-affiliated investigator by following the instructions here: https://www.nhlbiwgs.org/new-user-instructions. TOPMed-affiliated investigators will be granted access to the TOPMed website that contains the proposal template and a guide for submitting new TOPMed paper proposals.

Materials/Methods

VTE cases: ~1,100​​​
Ischemic stroke cases: ~4,000​
Hemorrhagic stroke cases: ~900
Non-cases (controls/cohort): ~5,100

Note: for VTE and the two stroke classes, these are essentially "all" currently available WHI cases with adjudicated outcomes, DNA (or buffy) availability, and dbGaP eligibility. The non-cases will be selected roughly as follows:

• Non-cases will be in proportion to HT trial (per arm) and ethnicity among the combined case groups.
• Within each (HT x ethnicity) stratum controls will be selected "fairly random", but preference will be given to WHI participants with measured biomarkers and WHI participants with already extracted DNA. (Details in the AS564 participant selection document).​

Sampling weights: Sampling weights, to be used in inverse probability weighted analyses, were constructed to relate the WHI TOPMed samples to the larger WHI cohort. Variables controlled for in this analyses were race/ethnicity, age, whether the participant was in the HT trials, and whether the person had experienced an MI, a stroke, or a VTE as of September 2019.

Whole genome sequencing (WGS) at Broad. WGS coverage performed at ~30X. Sequencing is complete and transferred to dbGaP.

CHIP: Somatic mutations with leukemogenic potential may confer selective advantages leading to clonal expansion, a phenomenon termed 'Clonal Hematopoiesis of Indeterminate Potential' (CHIP). To identify high-fidelity somatic mutations in blood-derived DNA in TOPMed, TOPMed WGS data were analyzed with the GATK MuTECT2 somatic variant caller using previously described methods (Jaiswal, NEJM 2014) to identify CHIP carriers on the basis of a prespecified list of leukemogenic driver mutations in genes known to promote clonal expansion of hematopoietic stem cells. The CHIP dataset is comprised of two files, TOPMed_CHIP_calls and TOPMed_CHIP_variants.

Telomere length: The TelSeq method (Ding et al, NAR 2014) was used to perform telomere length estimation on the TOPMed WGS data. TelSeq calculates an estimate of individual TL using counts of sequencing reads containing a fixed number of repeats of the telomeric nucleotide motif TTAGGG. These read counts are then normalized according to the number of reads in the individual with between 48% and 52% GC content, to adjust for potential technical artifacts related to GC content.