Skip to content

fix(gene burden): normalise NaN effect sizes to null #107

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
vivienho wants to merge 1 commit into
base: main
Choose a base branch
from
Open

fix(gene burden): normalise NaN effect sizes to null #107

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 7 additions & 0 deletions pts/src/pts/pyspark/gene_burden.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -452,6 +452,10 @@ def process_genes_and_health_gene_burden(
[_process_gnh_additive(st9_df), _process_gnh_meta(st13_df), _process_gnh_recessive(st15_df)],
)

# Some source rows encode a missing effect size as the string "NaN", which casting to double turns
# into a float NaN rather than a null. Normalise to null so beta/oddsRatio behave as missing.
gh_df = gh_df.withColumn('effect', f.when(f.isnan('effect'), f.lit(None)).otherwise(f.col('effect')))

# WARNING: some meta-analysis p-values underflow to 0.0 (inf LOG10P). Substitute the minimum non-zero
# p-value so they pass validation instead of being dropped.
gh_df = _substitute_zero_pvalues(gh_df, 'pValue', 'Genes & Health')
Expand Down Expand Up @@ -960,6 +964,9 @@ def _process_brava_granular(df: DataFrame) -> DataFrame:
.withColumn('BETA Burden', f.col('BETA Burden').cast('double'))
.withColumn('SE Burden', f.col('SE Burden').cast('double'))
)
# SKAT/SKAT-O rows report no effect size; the source encodes this as the string "NaN", which the
# cast above turns into a float NaN rather than a null. Normalise to null so it behaves as missing.
df = df.withColumn('BETA Burden', f.when(f.isnan('BETA Burden'), f.lit(None)).otherwise(f.col('BETA Burden')))
df = _substitute_zero_pvalues(df, 'Pvalue', 'BRaVa')

ancestry_group = f.col('meta analyzed')
Expand Down
Loading