Code

Feature cv training#55

Code

Merged

bfclarke merged 93 commits intomainPMBio/deeprvat:mainfrom feature-cv-trainingPMBio/deeprvat:feature-cv-trainingCopy head branch name to clipboardFeb 23, 2024

+2,151-326

Commits

Commits on Nov 20, 2023

• performance optimizations

  bfclarke

  committed
  6761a90

Commits on Nov 22, 2023

• train multiple repeats on single node in parallel

  bfclarke

  committed
  684eb23

Commits on Nov 23, 2023

• bug fix

  bfclarke

  committed
  4ad34d1

Commits on Nov 24, 2023

• fix bug in indexing when subset_samples() removed something

  bfclarke

  committed
  fffe0b4

• sleep between jobs; stop if any job fails

  bfclarke

  committed
  8fd60b7

Commits on Nov 27, 2023

• format with black

  bfclarke

  committed
  d121eff

Commits on Nov 28, 2023

• bug fixes

  bfclarke

  committed
  93c8227

• add test for MultiphenoDataloader

  bfclarke

  committed
  a5d9a1f

• update environments

  bfclarke

  committed
  5bdeea5

• uncomment rules

  bfclarke

  committed
  fe1680d

• bug fixes

  bfclarke

  committed
  58db4c4

Commits on Nov 30, 2023

• subset samples in training_dataset rule

  bfclarke

  committed
  72c8fc7

• example config.yaml

  bfclarke

  committed
  5dd6f5e

• use gpu queue for compute_burdens

  bfclarke

  committed
  b985e77

Commits on Dec 14, 2023

• bugfix since dask reading didn't work any more

  HolEv

  committed
  47e306d

Commits on Dec 16, 2023

• allow evaluation of all repeat combinations

  HolEv

  committed
  3f27d09

Commits on Dec 17, 2023

• allow analysis of each n_repeats and for all repeat combinations

  HolEv

  committed
  d0a0d7b

Commits on Dec 20, 2023

• option to provide burden file

  HolEv

  committed
  1e9587a

• allow seed gene alpha to be defined in config

  HolEv

  committed
  37efd46

• change sorting order to get the best model

  HolEv

  committed
  e488f4a

• adaptations to analyze multiple repeats and use script wo seed genes

  HolEv

  committed
  0d4833f

• allow to provide a sample file and do separate indexing for pheno and geno to ensure indices are correct

  HolEv

  committed
  1ba2619

• automatize generation of figure 3 (associations & repliation)

  HolEv

  committed
  b04d308

• generate cv splits with related samples in the same split

  HolEv

  committed
  07dbde4

Commits on Jan 3, 2024

• average burdens

  HolEv

  committed
  c4f90fe

• average burdens

  HolEv

  committed
  6bfc4dd

Commits on Jan 11, 2024

• cross-validation like trainign

  HolEv

  committed
  98a0b8c

• add missing cv_utils

  HolEv

  committed
  e10ec47

• write average burdens or each combination to single zarr file to avoid zarr issues

  HolEv

  committed
  b43b7e5

• add logging information

  HolEv

  committed
  7acc840

Commits on Jan 16, 2024

• make maf column a param

  HolEv

  committed
  ed0ebfd

• add logging

  HolEv

  committed
  18a9e3c

• pipeline replictaion and plotting

  HolEv

  committed
  08603b7

• evaluate all repeat combis with and without seed genes

  HolEv

  committed
  4fdb112

• update lsf.yaml

  HolEv

  committed
  e7eb810

Commits on Jan 18, 2024

• small updates

  HolEv

  committed
  ebbd7d0

• per-gene pval aggregation

  HolEv

  committed
  c94466e

• aggregate pval per gene

  HolEv

  committed
  b489f5b

Commits on Jan 29, 2024

• bugfix- only load burdens if not skip burdens

  HolEv

  committed
  90be746

• logging info

  HolEv

  committed
  291904f

• updates and fixes

  HolEv

  committed
  425036c

• load burdens only for genes analysed in current chunk to save memory

  HolEv

  committed
  f7c64a1

Commits on Feb 12, 2024

• small changes to pipeline

  HolEv

  committed
  3772651

• standardizing/qt-transform of combined test set x/y arrays

  HolEv

  committed
  74c905b

• my_quantile_transform for numpy arrays

  HolEv

  committed
  24c3475

• bugfix

  HolEv

  committed
  c150ba7

• remove unnecessary code

  HolEv

  committed
  59b5ed3

• remove unnecessary wildcards

  HolEv

  committed
  296c6fe

• make averaging part of associate.py

  HolEv

  committed
  7fe5dc7

• allow seed genes/baselines to be missing (to allow assoc. testing for non-training phenotypes)

  HolEv

  committed
  618f7fc

• updates

  HolEv

  committed
  17a104b

• gene-specific common variant covariates for conditional analysis

  HolEv

  committed
  bfecab6

Commits on Feb 15, 2024

• bugfix

  HolEv

  committed
  6d8a935

Commits on Feb 16, 2024

• post-hoc conditioning on common variants

  HolEv

  committed
  bc13830

• restructure pipelines

  HolEv

  committed
  d105fc3

Commits on Feb 19, 2024

• removing redundant options

  HolEv

  committed
  d303fe7

• add cv_utils cli

  HolEv

  committed
  15efe9d

• simplify script (only evaluate one repeat combi/average burdens); aggregate baseline pvalues; make bonferroni correction default

  HolEv

  committed
  64974bf

• removal of redundant wildcards, updates and fixes

  HolEv

  committed
  dd0852c

Commits on Feb 20, 2024

• bugfixes

  HolEv

  committed
  24e8ce2

• baseline discoveries only required for training phenotypes

  HolEv

  committed
  8ca566c

• remove not needed code

  HolEv

  committed
  85147f9

• update configs

  HolEv

  committed
  07a6c83

• formatting

  HolEv

  committed
  c22dcc5

Commits on Feb 21, 2024

• manually merge changes from feature-regenie to account for gene-specific annotations

  HolEv

  committed
  011717a

• allow different sample orders in phenotype_df and genotypes.h5

  HolEv

  committed
  43bdba8

• change sample ids to be bytes as it is in the real data

  HolEv

  committed
  6d76d7c

• update pipelines

  HolEv

  committed
  6d06cf5

• update gitignore

  HolEv

  committed
  17dfb1f

• pipeline updates

  HolEv

  committed
  51caa5f

• manually update github actions to be like master

  HolEv

  committed
  33f35a1

• bug fixes

  HolEv

  committed
  06a9274

• checkout tests from master

  HolEv

  committed
  32ca35f

• make phenotype indices string as they are in real data

  HolEv

  committed
  c47c30f

• 'add gene_id' column

  HolEv

  committed
  d0aaa63

• manually merge with master so tests can pass

  HolEv

  committed
  4c43dc6

• bugfixes

  HolEv

  committed
  0b18840

• use gene_id column instead of gene_ids

  HolEv

  committed
  3a9ea91

• pipeline updates and fixes

  HolEv

  committed
  7938ad5

Commits on Feb 22, 2024

• update test config

  HolEv

  committed
  cb6a1fc

• adding age2 and age_sex to example data

  HolEv

  committed
  fd3db26

• update config

  HolEv

  committed
  cb5ddd1

• set tests folder to main version

  HolEv

  committed
  3c62850

• checkout preprocssing files from main

  HolEv

  committed
  5ac374e

• checkout from main

  HolEv

  committed
  d1c3d5f

• manually merge sample_id changes from main

  HolEv

  committed
  4f1c61c

• fixing merge conflicts with main

  HolEv

  committed
  6acdcf4

• pipeline bugfixes and renamings

  HolEv

  committed
  e42b478

• fixup! Format Python code with psf/black pull_request

  

  PMBio

  authored and

  PMBio

  committed
  071479b

• remove gene_ids column

  HolEv

  committed
  89bac25

• Merge branch 'feature-cv-training' of https://github.com/PMBio/deeprvat into feature-cv-training

  HolEv

  committed
  97f3a0b

Commits on Feb 23, 2024

• integrating suggested PR changes

  HolEv

  committed
  a0c1583

• fixup! Format Python code with psf/black pull_request

  

  PMBio

  authored and

  PMBio

  committed
  f86de06