| Age | Commit message (Expand) | Author |
| 7 days | api: scans point at the class counts behind their own peak | Pjotr |
| 7 days | docs: four analysis gotchas, with DOL trait 10003 as the worked example | Pjotr |
| 8 days | genotypes: a trait summarised per genotype class at one marker | Pjotr |
| 8 days | genes: Wikidata aliases, cached in gene-aliases.lmdb for seven days | Pjotr |
| 8 days | probesets: cross-dataset lookups by probeset id and by gene symbol | Pjotr |
| 8 days | api: every response suggests where to go next | Pjotr |
| 8 days | genotypes: call frequencies, marker names, and a location per marker | Pjotr |
| 8 days | compute/qtl2: multiparent crosses, by name | Pjotr |
| 8 days | compute: R/qtl2 as a third scan, POST /compute/qtl2 | Pjotr |
| 8 days | docs: the skill says fitqtl fits covariates | Pjotr |
| 8 days | compute: fit covariates in /compute/fitqtl too | Pjotr |
| 8 days | compute: add /compute/fitqtl for effect sizes at named markers | Pjotr |
| 8 days | covariates: refuse a mostly-numeric attribute, and teach the skill to use them | Pjotr |
| 8 days | api: normalize trait values, and winsorise the tails, before mapping | Pjotr |
| 9 days | compute: report GEMMA's effect size next to the score | Pjotr |
| 9 days | compute: let GEMMA name the trait, so its plot can be titled | Pjotr |
| 9 days | move dev SKILL.md | pjotr |
| 9 days | SKILL | pjotr |
| 9 days | Moved to SKILL file and updated | pjotr |
| 9 days | api: name the caller on the response, so the log can say who | Pjotr |
| 9 days | plot: say which trait a plot is, not just draw it | Pjotr |
| 9 days | compute: gate GEMMA from 50 individuals, R/qtl from 1 000 | Pjotr |
| 9 days | api: describe case attributes, and gate large computations only | Pjotr |
| 9 days | docs: match the server by argv[0], not by a substring of its command line | Pjotr |
| 9 days | docs: a restart that reports success and changes nothing | Pjotr |
| 9 days | compute: say what each covariate note is actually about | Pjotr |
| 9 days | compute: refuse GEMMA permutations instead of ignoring them | Pjotr |
| 9 days | docs: tell users (and operators) to refetch the instructions when the build m... | Pjotr |
| 9 days | compute: fit covariates in GEMMA, and share them with R/qtl | Pjotr |
| 9 days | compute: gate ravanan at one job, and name the binary that failed | Pjotr |
| 9 days | compute: gate R/qtl at eight concurrent scans, one per caller | Pjotr |
| 10 days | compute: one gemma job at a time, per caller, via #[compute_gate] | Pjotr |
| 10 days | docs: what the last two commits changed, and the dev/production split | Pjotr |
| 10 days | http: accept bodies up to 64 MiB, not axum's 2 MiB default | Pjotr |
| 10 days | db: recover from a connection the server closed, and report the database | Pjotr |
| 10 days | compute/rqtl: fit covariates, by case-attribute name or as numbers | Pjotr |
| 10 days | case-attributes: one attribute for one trait, in that trait's sample order | Pjotr |
| 10 days | ai-instruction: a section on case attributes | Pjotr |
| 10 days | plot: take a posted result, and never re-run the mapping | Pjotr |
| 10 days | plot: draw a mapping, and suggest it from the mapping | Pjotr |
| 10 days | compute: suggest 100 permutations after a scan, and say what GEMMA can do | Pjotr |
| 10 days | compute: suggest a PheWAS at the peak, and the plot that follows | Pjotr |
| 10 days | phewas: every trait peaking within a window of one position | Pjotr |
| 10 days | version: document the provenance fields, and assert they agree | Pjotr |
| 10 days | genotypes: expose the metadata, never the genotype calls | Pjotr |
| 10 days | genotypes: list a group's genotype files, and map by group + version | Pjotr |
| 10 days | docs: a new genotype file needs no restart; only code and env do | Pjotr |
| 10 days | docs: parents and F1s are excluded by default, measured both ways | Pjotr |
| 10 days | docs: name the real genotype files, panels and timings everywhere | Pjotr |
| 10 days | docs: which genotype file is current, and three real mappings measured | Pjotr |