| Age | Commit message (Expand) | Author |
| 7 days | docs: which reference genome the coordinates are on, and what the index covers | Pjotr |
| 7 days | docs: /compute/gemma/ravanan reports pve too, from its archives | Pjotr |
| 7 days | gemma: pve on the ravanan/slurm path, read out of each archive | Pjotr |
| 7 days | gemma: the pve note carries every chromosome, and the log carries the numbers | Pjotr |
| 7 days | bulklmm: report Julia's error line, not its last line | Pjotr |
| 7 days | next: every link a client is offered is a call it can make | Pjotr |
| 7 days | docs: the two engines agree on heritability -- measured, not asserted | Pjotr |
| 7 days | docs: how pve gets deployed, and what the deployed box runs | Pjotr |
| 7 days | gemma: report GEMMA's own pve for the null model | Pjotr |
| 7 days | docs: GEMMA's pve is one flag away, not a dead end | Pjotr |
| 7 days | docs: -pve is untested, not refuted | Pjotr |
| 7 days | docs: how far the GEMMA pve route got | Pjotr |
| 7 days | docs: why the two engines' h2 disagree, and the three measurements that settl... | Pjotr |
| 7 days | genes: the alias reply carries the entity's label and its gene link-outs | Pjotr |
| 7 days | docs: the gene link-outs GN2's trait page offers, checked and tableted | Pjotr |
| 7 days | plot/bulklmm: the 0.05 and 0.01 lines, never 0.10 unless asked | Pjotr |
| 7 days | docs: say that /plot/gemma draws no line unless one is asked for | Pjotr |
| 7 days | plot/gemma: no reference line by default; Bonferroni is one more labelled line | Pjotr |
| 7 days | plot/gemma: draw thresholds computed elsewhere, labelled with their source | Pjotr |
| 7 days | bulklmm: report the variance components, and say what a boundary means | Pjotr |
| 7 days | bulklmm: BULKLMM_BLOCKS as a config option, default 4 | Pjotr |
| 7 days | gemma: the permutation refusal names both routes, including running GEMMA itself | Pjotr |
| 7 days | docs: what BulkLMM models, what its cores do, and how to package it | Pjotr |
| 7 days | bulklmm: put the whole driver summary in the reply | Pjotr |
| 7 days | bulklmm: exact marker-block parallelism, and what the cores are actually doing | Pjotr |
| 7 days | bulklmm: an LMM scan that can be asked for thresholds | Pjotr |
| 7 days | qtl2: name the genotype model, and say what the conversion dropped | Pjotr |
| 7 days | qtl2: an option to keep heterozygous calls, to test what dropping them costs | Pjotr |
| 7 days | traits: genotypes at a confirmed peak, and the plot of them | Pjotr |
| 8 days | uploads: every engine maps an upload by name; same name replaces | Pjotr |
| 8 days | uploads: map an uploaded genotype by name, and clean up after it | Pjotr |
| 8 days | uploads: a genotype matrix that updates an existing population | Pjotr |
| 8 days | baskets: clear orphan rows on open, so an affected file heals | Pjotr |
| 8 days | matrices: upload new traits and covariates as one named table | Pjotr |
| 8 days | baskets: every revision kept, files named for the user, all four engines | Pjotr |
| 8 days | baskets: a user's own column of values, in their own SQLite file | Pjotr |
| 8 days | api: scans point at the class counts behind their own peak | Pjotr |
| 8 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: give the Wikidata call 30 s, not 20 | 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 |
| 9 days | gemma: quote GEMMA's own error, not the wrapper's banner | Pjotr |
| 9 days | fix: DOL maps again on both engines, and the four stale tests | Pjotr |
| 9 days | qtl2: the engine script and its packaging | Pjotr |
| 9 days | docs: the skill says fitqtl fits covariates | Pjotr |