pirl-unc / hitlist / 24699890996
80%

DEFAULT BRANCH: main
Ran
Jobs 1
Files 21
Run time 1min
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst
coverage: 48.105% (+0.8%) from 47.309%
24699890996

push

github

web-flow 
v1.13.0: Comprehensive Bekker-Jensen 2017 ingest — full Fig 1b design matrix (closes #89) (#103)

* v1.13.0: Comprehensive Bekker-Jensen 2017 ingest — full Fig 1b design matrix

Expands the bulk proteomics peptide + protein index to cover every
experimental arm of Bekker-Jensen 2017 (PMID 28591648, PXD004452)
present in the deposit. Prior version shipped only 46-fraction tryptic
cell-line runs + 39-fraction non-tryptic HeLa. This commit adds HeLa
tryptic runs at 14 / 39 / 46 / 70 fractions, plus TiO2 phosphopeptide-
enriched runs at 12 fractions (pH8, pH10) and 50 fractions.

Schema (row-level columns on bekker_jensen_2017_peptides.csv.gz):
- digestion_enzyme  Trypsin/P, Chymotrypsin, GluC, LysC
- n_fractions_in_run  14 / 39 / 46 / 70 (plus 12/50 for TiO2 arms)
- enrichment  "none" or "TiO2"
- modifications  MaxQuant Modifications column
- n_replicates_detected  count of replicates that observed the peptide

Protein-level CSV extended analogously; abundance_percentile is now
ranked within each experimental arm (cell_line, enzyme, n_fractions,
enrichment) because absolute log2 intensities are only comparable
within one arm.

Peptide index grows from 1.3M to 2.03M rows; protein index from 71K
to 151K rows. Total parquet footprint ~45 MB (up from ~32 MB).

Loader changes (hitlist.bulk_proteomics):
- load_bulk_peptides() and load_bulk_proteomics() accept new kwargs:
  digestion_enzyme=, n_fractions_in_run=, enrichment=.
- enrichment defaults to "none" — baseline detectability queries never
  mix in TiO2 phospho-enriched rows unless explicitly requested with
  enrichment="TiO2" or enrichment=None. Phosphopeptides fly and
  fragment differently; mixing them into a flat detectability prior
  would bias downstream MS-bias models toward phospho-compatible
  sequences.

Sanity checks:
- C-term specificity per enzyme matches cleavage rules: Trypsin/P
  98.7% K/R; LysC 98.0% K; GluC 96.9% E/D (bicarbonate buffer spec
  "GluC;D.P"); Chymotrypsin+ 92.7% F/W/Y/L/M.
... (continued)

1472 of 3060 relevant lines covered (48.1%)

0.48 hits per line

Coverage Regressions

Lines Coverage ∆ File
89
49.47
 1.67% builder.py
Jobs
ID Job ID Ran Files Coverage
1 24699890996.1 21
48.1
 GitHub Action Run
Source Files on build 24699890996