Normalization 1.1.2 reproduces all 109,191 bins of the 1000 Genomes
Omni2.5 GTCs (AutoCall 1.6.2.2) with 0 ULP difference, reserved fields
included; 1.2.0 and 1.1.2 are bit-identical to the no-FMA gtc2vcf oracle
on 36 GSA samples. AWS run: 33 min, ~$0.15, no resources left.
- fit.py: exact and binned nearest-neighbour searches use one distance
matrix per searched bin (first-minimum rule kept); fit 2.6 s -> 1.4 s
per GSA sample, still bit-identical to the oracle
- scripts/validate_p3_large.py: refits 1.2.0 and 1.1.2 transforms for
every vendor GTC and reports ULP distances per field
- cloud trial runs scripts/cloud/tasks/<TASK>.sh (TASK=p3 default;
p1p2 = previous validation + ingest)
- signal/normalization/numerics.py: vendor percentile, trimmean, MAD sigma
(old/new), bisquare robust fits, float32/float64 typing and sequential
sums reproduced exactly (ported from gtc2vcf, MIT)
- signal/normalization/fit.py: outlier removal, offset, rotation, shear,
scale, nn12/rr12, single-channel model for small Infinium I bins,
exact and binned nearest-neighbour sweeps; vendor quirks marked QUIRK
- signal/normalization/apply.py: affine transform and theta/R
- pipeline/normalize.py + CLI `simurg normalize`
- oracle GTCs rebuilt with -ffp-contract=off (aarch64 gcc fuses a*b+c by
default, shifting the last bits); the old FMA oracle differed by <=4 ULP
Result: transforms bit-identical to the oracle for all 36 GSA samples x
48 bins, for both normalization versions (3456/3456 bins, 0 ULP).
Raw X/Y and controls assembled from IDATs equal the vendor GTC arrays for
2138/2141 samples x 2.45M loci; 3 single-locus differences are recorded
as K-009 (probable per-channel missing rule in AutoCall 1.6.2.2). Channel
order X=Red confirmed (U1). `simurg ingest` processed all 2141 samples in
19 min (I/O-bound). AWS run cost ~$0.34 and left no resources.
- tests/oracle/test_p2_raw_xy.py: IDAT-assembled raw X/Y and controls
equal the arrays in oracle GTCs
- validate_p1_large.py: phase-P2 comparison against vendor GTCs with
diagnostics (channel swap, bead-count disagreements)
- cloud trial: also runs `simurg ingest` on all samples (time/memory),
disk 250 GB
All 2141 vendor-written GTC v3 files match BeadArrayFiles field by field
and re-encode byte-identically; all IDAT pairs parse; the 2.45M-locus BPM,
dbv-7 EGT and CSV manifest agree with BeadArrayFiles and each other.
Run on a self-destructing AWS spot instance (42 min, ~$0.20).
scripts/cloud/trial.sh launches one spot instance that validates the
parsers on the 1000 Genomes Omni2.5 set (2141 samples, vendor GTCs)
streamed from the public S3 bucket, uploads the summary and terminates
itself. Cost guards: spot price cap, hard time limit (shutdown =
terminate), EXIT-trap shutdown, CloudWatch idle alarm that terminates
from outside, disk deleted with the instance, monthly budget alert, and
an `audit` command listing anything still billable.
scripts/validate_p1_large.py compares every GTC with BeadArrayFiles,
checks byte-identical round-trips, IDAT pairs and BPM/EGT/CSV.
- pyproject (uv/hatch, ruff, mypy strict, pytest markers), MIT LICENSE,
THIRD_PARTY_NOTICES with verbatim BSD-2 (BeadArrayFiles) and MIT (gtc2vcf)
- CI workflow for Gitea/GitHub Actions: lint, type check, unit tests
- fetch_data.sh p0 now fetches the GSA v4 PGx demo IDATs (the GDA "Demo 36"
bundle is a GenomeStudio project), unpacks archives, fixes pipefail bug
- make_oracle_gtcs.sh builds bcftools + gtc2vcf and writes oracle GTCs
- plan §17: corrected P0 dataset table
Research notes R1-R4, documentation-study files A-E (source of the
functional spec's `src:` tags) and PGx coverage CSVs, rewritten to use
neutral naming and current project decisions (plan §2).
Phase 0 deliverables: full technical specification and roadmap
(SIMURGARRAY_MASTER_PLAN.md), CLAUDE.md, bootstrap script for reference
open-source repos and tiered dataset fetch script with checksums.
kemal