pirl-unc / hitlist / 34256653995
84%

DEFAULT BRANCH: main
Ran
Jobs 1
Files 30
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: 83.474%. Remained the same
34256653995

push

github

web-flow 
Correct four studies whose sample curation source verification disproved (#436) (#441)

Each entry below was internally consistent and wrong, and each would have
passed a plausibility check. Three failure modes recur.

A plausible cell line substituted for the real one. PMID 34497125 carried
A375 and trametinib; the paper says SKMEL5 and binimetinib throughout and
mentions neither of the others, and the PXD024917 file names agree. The
damage was not the label: the per-sample mhc was A375's genotype attached
to a study that never used A375. It is withdrawn rather than substituted,
since neither the paper nor cell_lines.yaml records SKMEL5's typing.

A perturbation axis collapsed. PMID 34129938's arms are all IFN-gamma
treated and differ by doxycycline induction and dTAG-13 degradation; one
"unperturbed MC38" entry asserted the opposite for every quantified arm,
on the engineered idAdpgkG derivative rather than the parental line. PMID
39438697's four THP-1 conditions are a 2x2 of TAP1 knockout by Mtb
infection, named verbatim in the authors' conditions_table_TAP.csv; two
arms cannot express that. Its primary hMDMs were absent and its
splenocytes, which the paper pulses with Alg8 peptide, were curated
unperturbed. Both studies now reach apm_genes_perturbed (ifn_gamma,
tap1) -- a collapsed axis is invisible to every downstream filter.

A citation inverted. PMID 27846572's C1R note sourced the data to
Bassani-Sternberg 2015. The paper cites C1R to reference 5 (Caron 2015)
and the fibroblasts to reference 6 (Bassani-Sternberg 2015), exactly
backwards.

That study also carried JY and HeLa samples the paper never mentions.
They were not inert. T2's 111 rows were being attributed to JY by
allele_exact -- JY's curated HLA-A*02:01 matched them at the
highest-confidence tier -- so a TAP-deficient hybridoma was labeled an
EBV-LCL, while T2 itself was absent from the curation. Removing the two
phantom arms and adding T2 moves exactly those 111 rows, the only
attrib... (continued)

6829 of 8181 relevant lines covered (83.47%)

0.83 hits per line

Jobs
ID Job ID Ran Files Coverage
1 34256653995.1 30
83.47
 GitHub Action Run
Source Files on build 34256653995