Omevra · NGS Pipeline Atlas — proof

NGS Pipeline Atlas — evaluations

Name: NGS Pipeline Atlas
Brand: Omevra

Evaluation v1: two end-to-end runs of the 33-sample cohort, the gate record, and the run reports.

Evaluation v1 — two runs, 33 samples

Version label: Evaluation v1 — two end-to-end runs, 33-sample cohort. One Nextflow workflow (main.nf) processes the ENA study PRJNA234495 E. coli K-12 cohort (reference NC_000913.3) through fetch (MD5-verified), a read-pair sanitize stage, FastQC/fastp, BWA-MEM alignment, per-sample QC gates, bcftools calling and a cohort report. Gates are fixed before the run: mean coverage ≥ 10.0×, reads mapped ≥ 80%, Q30 bases ≥ 70%. A sample that fails a gate has its BAM removed inside the gate step and never reaches calling. The evaluation ran the identical 33-sample sheet twice, in fully separate work and output directories — run 2 in an isolated launch directory, so no task result could be reused from run 1's cache — with the locked environment in environment-explicit.txt.

Results — identical in both runs

33Samples attempted

27Passed all QC gates to variant calling

6Stopped at the ≥10× coverage gate

130,614Variant records across passed samples — 81,810 unique sites

Passed all three gates — went on to variant calling27

Stopped at the coverage gate6

The six stops — mean coverage against the 10.0× gate

Sample	Mean coverage	Coverage gate	Note
BADINPUT01 (deliberate bad input)	0.01×	≥ 10.0×	deliberate bad input
SRR1121841	8.82×	≥ 10.0×	genuine cohort sample
SRR1121895	9.03×	≥ 10.0×	genuine cohort sample
SRR1121918	9.33×	≥ 10.0×	genuine cohort sample
SRR1121923	9.70×	≥ 10.0×	genuine cohort sample
SRR1121929	9.99×	≥ 10.0×	genuine cohort sample

All six stops had ~98% of reads mapped and Q30 of 96% — coverage was the deciding metric. The gate log line for the deliberate bad input, verbatim and identical in both runs: [GATE] BADINPUT01 STOPPED reason: mean_coverage=0.01x < gate 10.00x (mean_coverage=0.01x pct_mapped=97.42 q30_pct=96.00)

Reproducibility

Cohort reports byte-identical between run 1 and run 2 — MD5 822aad36… for both.
Per-sample variant tables: 27 of 27 byte-identical.
Sanitize statistics identical in both runs: 27,175,960 read pairs processed, 27,071,276 kept, 104,684 corrupt pairs dropped (0.39%).
Gate decisions and metrics identical for all 33 samples, including all six stops.

Per-stage timing — median task time, run 1 trace

Stage	Tasks	Median task time
FETCH	33	6 s (local mirror)
SANITIZE	33	66 s
FASTQC	33	11 s
FASTP	33	20 s
ALIGN	33	44 s
GATE	33	4 s
CALL	27	24 s
REPORT	1	1 s

Downloads — the run reports

RESULTS.md — the full results document behind every figure on this page
cohort_report_run1.txt — run 1 cohort report
cohort_report_run2.txt — run 2 cohort report, byte-identical to run 1
summary_run1.tsv — run 1 per-sample summary table
summary_run2.tsv — run 2 per-sample summary table
gate_log_run1.txt — run 1 gate log, all samples
gate_log_run2.txt — run 2 gate log, all samples

Each file is copied byte-identical from the project's run outputs at build time; the build verifies the copies.

Product: NGS Pipeline Atlas.