Sprint 15 Task List

[cristinaetrv]

For August 8 Demo
Goal: Finish improvements that ADRC are looking for

• Share only some samples/subjects
• PRS with Dave's results
• Summarize APOE clinical genotypes (get APOE E4/E4, E2/E2, etc per sample)

Annotation

• Give ETA to Dave/Greg Gibson/IBDGC on annotation changes:

Proteomics
Goals: Get user story, talk to more users (UI standpoint)

• Follow Eric Dammer's pipeline using DA denoiser @akotlar
• Replace Eric Dammer's network analysis with SPPCA and compare @akotlar
• Support pulling prot data from Bystro cluster @akotlar
• Document pulling prot data from Bystro cluster @akotlar
• Start contacting more proteomicists for UI ideas @akotlar
• Read papers on multi-modal analysis (pQTLs etc) @cristinaetrv @akotlar
• Determine what Bystro multi-modal analysis will look like by 7/29 @cristinaetrv @akotlar

Subject mapping

• Select fields to correspond with Bystro analysis module conventions
• Add ability to specify whether variables like phenotypes are continuous or categorical @akotlar

PRS

• (Backlog) Imputation method, write python bindings and have Rust code compile and be callable from python @akotlar
• (Backlog)Take in ancestry PCs/Take in GWAS summary statistics as PRS-CS covariates @akotlar
• Harmonized summary stats for IBD in our format @cristinaetrv 7/25/24
• Lift over harmonized summary stats to hg19 @cristinaetrv 7/25/24
• Rerun Dave's version of PRS with IBD stats to compare with IBD on our version @cristinaetrv 7/25/24
• Calculate odds ratios using threaded through covariate information @akotlar 7/26/24
• UI improvements for advanced settings @akotlar

Ancestry V2 (post subject mapping integration):

• Webapp: Expose ability to set expected superpopulation and compare in the ancestry score section
• Webapp: flag samples that have different superpopulation than specified covariate superpopulation

Infrastructure

• Investigate how difficult integrating existing imputation method as an optional step before annotation @akotlar
• Add test coverage for submission functions, progress message updates (queue server responding to submitting events) @akotlar
• Add user management ie change password, add S3 bucket, delete account, etc
• Documentation on APIs for GIN deployment @akotlar 7/29/24

POE

• Mike E lab (Taylor) runs on UK Biobank
• Finish BCA (bias corrected method) @austinTalbot7241993 8/2/2024
• Experiments and benchmarking after we get feedback from Mike E @austinTalbot7241993
• Jupyter notebook tutorial for how to run the method @IlhaH 8/7/24

Hypothesis testing

• Write up draft of paper @IlhaH @austinTalbot7241993
• Reading papers about hypothesis testing @IlhaH 7/26/24
• Converting hypothesis testing methods from R to python @IlhaH
• Share overleaf link of draft @austinTalbot7241993 7/23/24
• Talk to Xiucai about where paper should be submitted, which methods should we incorporate, how to test behavior 7/29/2024
• Send Xiucai summary documents 8/6/24

Covariance Matrix Estimation Method

• Write up description of method, benchmarking information, document how to use (paper format) as a supplement for POE @IlhaH 8/8/24

Future work (next sprint or later)

• Run fine-grained compute for Bystro cluster (Ray) in Q3
• Enable VCF combine by sample in Q3

Grant drafts

• Specific aims (1 page) @akotlar @cristinaetrv
• Project narrative (3 sentences) @cristinaetrv
• Project summary (30 lines) @cristinaetrv
• Research strategy (6 pages) @cristinaetrv
• Budget (no stated limit) @akotlar
• Facilities and other resources (no stated limit) @cristinaetrv
• Resource sharing plans (no stated limit) @cristinaetrv