search

fpbarthel / GLASS

GLASS consortium
MIT License
37 stars 13 forks source link

Review variant calls #85

Closed fpbarthel closed 5 years ago

fpbarthel commented 5 years ago

Related to https://github.com/TheJacksonLaboratory/GLASS-WG/issues/84

Additionally, we should manually review a dozen or two discrepancies across several other genes. I generated this list as follows:

SELECT gt.aliquot_barcode, v.chrom, v.start, v.end, v.ref, v.alt, gene_symbol, hgvs_c, variant_classification, variant_type, gt.called, gt.read_depth, gt.alt_count,
ROUND(CAST(gt.alt_count AS DECIMAL)/CAST(gt.read_depth AS DECIMAL),2) AS vaf,
(ROUND(CAST(gt.alt_count AS DECIMAL)/CAST(gt.read_depth AS DECIMAL),2) > 0.1 and read_depth > 3) AS manual_call
FROM analysis.snvs v, analysis.snv_genotypes gt
WHERE gt.chrom = v.chrom AND gt.start = v.start AND gt.end = v.end AND gt.alt = v.alt
AND gene_symbol IN ('TP53','ATRX','RB1','EGFR','PTEN','TERT','NF1','CIC','FUBP1','PDGFRA','PIK3CA','PIK3R1')
AND variant_classification != 'Intron' 
AND (called = true OR (ROUND(CAST(gt.alt_count AS DECIMAL)/CAST(gt.read_depth AS DECIMAL),2) > 0.1 and read_depth > 3) = true)

Data: data-1541605302668.xlsx

fpbarthel commented 5 years ago

Tabled in favor of more pressing issues atm