# repoPR numberone-line descriptiontype (bug fix | speedup | tests/CI | tool; optional, default bug fix). tools/render_bio_prs.py fills status from the GitHub API. synthesizebio/pysynthbio 77 `predict_query` no longer writes its options into the caller's query dict EdisonScientific/labbench2 45 Leading-decimal answers like `.701` keep their value when graded EdisonScientific/labbench2 46 Report summaries keep metrics whose scores are all 0 snap-stanford/Biomni 372 FDA adverse-event summaries count deaths and hospitalizations from the report-level flags snap-stanford/Biomni 373 `query_openfda` sends its search, sort, count and skip parameters latchbio/latch 624 Glob patterns match the whole file name (`*.fq` no longer matches `a.fq.gz`) latchbio/latch 625 `generate-metadata` handles nf-core schema patterns that crashed it (55 to 124 of 146 pipelines) robocurve/inspect-robots 617 Rollouts record a copy of each executed action, so a reused policy buffer can't change scores robocurve/inspect-robots 618 Trial artifacts stay separate for scene IDs that differ only in case rowansci/rowan-python 19 Scan energy positions, explicit empty pKa element lists, and folder lookup past the first page seqeralabs/nf-metro 2172 Converted metro maps keep edges into dead-end processes in other sections seqeralabs/nf-metro 2173 Subworkflows with the same short name get separate sections instead of overwriting each other Arcadia-Science/agentic-pymol 18 get_model works against real PyMOL (Indexed was imported from the wrong module) Arcadia-Science/agentic-pymol 19 Relative file paths are resolved by PyMOL, as documented, not by the MCP server chaidiscovery/chai-lab 437 Merging A3M files from the same database keeps every file's hits adaptyvbio/protein-design-skills 10 The UniProt skill's PDB helper puts the chains, not the resolution, in its chains field adaptyvbio/protein-design-skills 11 The setup check uses modal token info, since modal token show does not exist rowansci/openconf 19 Heavy-atom RMSD searches symmetry mappings, so symmetric conformers compare as identical rowansci/openconf 20 random_seed=0 gives reproducible conformers instead of an unseeded run rowansci/steamroll 8 Transition-metal conversions keep input atom order and supplied coordinates chaidiscovery/chai-lab 438 Template lookup keeps chain names like 001 or NA as strings nf-core/tools 4619 create-params-file includes parameters defined at the schema's top level bug fix nf-core/tools 4620 Schema build keeps 0 and 0.0 defaults from the pipeline config bug fix Future-House/aviary 393 Tool servers build for tools with list parameters (JSON Schema "array") bug fix Arcadia-Science/glass-box-umap 34 Float64 Tensor input works in transform and compute_contributions, like NumPy input bug fix MultiQC/MultiQC 3736 A module set only in top_modules no longer runs a second time on all logs bug fix Future-House/paper-qa 1373 Text parsing compares page_size_limit with each line's length, not the line count bug fix Future-House/paper-qa 1374 Rebuilding an index with use_absolute_paper_directory keeps unchanged documents bug fix latchbio/scbench 8 scbench batch skips evals/manifest.json instead of counting it as a failed eval bug fix Arcadia-Science/glass-box-umap 35 py.typed marker so type checkers use the package's annotations (issue #18) tool Future-House/aviary 394 Tool servers pass a parameter's declared default when the request leaves it out bug fix ArcInstitute/state 283 Gene mapping uses the detected gene-name column, so genes named only there aren't dropped bug fix ArcInstitute/cell-eval 248 score_agg_metrics works when one normalization group is empty bug fix ArcInstitute/cell-eval 249 Pipeline metric settings no longer leak into other pipelines' defaults bug fix ArcInstitute/evo2 236 An interrupted shard merge no longer leaves a truncated checkpoint that later loads accept bug fix snap-stanford/Biomni 376 Drug recall search applies the requested classification filter bug fix snap-stanford/Biomni 377 Adverse-event search reports FDA API errors instead of 'no adverse events found' bug fix NVIDIA-BioNeMo/bionemo-recipes 1802 ESM-2 PEFT validation packing uses val_micro_batch_size bug fix EdisonScientific/labbench2 48 custom runner outputs crash after generation (OSError) bug fix EdisonScientific/labbench2 49 --retry-from ignores the original agent/mode; merges mix configurations bug fix ArcInstitute/cell-eval 250 sparse pseudobulk and centroids without densifying (3x faster, 4.8x less memory) speedup robocurve/inspect-robots 622 HTML report scans the frame directory once (16x faster, byte-identical) speedup snap-stanford/Biomni 378 database REST helper: timeouts and bounded GET retries bug fix Future-House/paper-qa 1375 reuse NumpyVectorStore matrix and norms across retrievals (70x warm, bit-identical) speedup sunpy/sunpy 8919 Fixes CROTA fallback for EUI maps bug fix scverse/scanpy 4531 fix: match expression layers before reusing marker statistics bug fix scverse/scanpy 4532 fix: keep legacy graph metrics usable without anndata.acc bug fix scverse/scanpy 4533 fix: derive PCA defaults from the selected representation bug fix scverse/scanpy 4534 fix: preserve gene pool alignment in score_genes bug fix latchbio/latch 626 launch: send false Nextflow flags and pick the File/Dir union variant by type bug fix latchbio/latch 627 file_glob: put a slash between remote_directory and the file name bug fix rowansci/openconf 21 hybrid: pick low-mode source seeds from the pool after seed insertion bug fix rowansci/openconf 22 pool: apply energy_window_kcal to the final selection bug fix rowansci/steamroll 9 to_rdkit: check transition-metal results against the supplied SMILES bug fix Arcadia-Science/agentic-pymol 21 types: drop slots from tool result dataclasses so FastMCP publishes output schemas bug fix materialsproject/pymatgen 4787 ZSLGenerator: cache super-lattice reductions, vectorize match check (3.3-4.8x, byte-identical) speedup materialsproject/pymatgen 4788 Fix GibbsComputedStructureEntry.from_dict for entries with a correction bug fix materialsproject/pymatgen 4789 Fix QuasiRRHO free-rotor interpolation using vibrational temperatures instead of wavenumbers bug fix materialsproject/pymatgen 4790 Fix MagneticStructureEnumerator crash when the ferromagnetic strategy is not used bug fix materialsproject/pymatgen 4791 Fix get_plot and plot_structures with two_theta_range=None bug fix materialsproject/pymatgen 4792 Fix XAS.stitch in XAFS mode when called on the EXAFS spectrum bug fix materialsproject/pymatgen 4793 Normalize data["volume"] per formula unit in InsertionVoltagePair bug fix materialsproject/pymatgen 4794 Fix get_dimensionality_cheon modifying the input structure when standardize=False bug fix materialsproject/pymatgen 4795 Return "intercalated ion" from get_dimensionality_cheon with larger_cell=True bug fix astropy/astropy 20894 Fix sigma_clip rejection masks for Quantity input bug fix astropy/astropy 20895 Honor zero sigma bounds in SigmaClip bug fix astropy/astropy 20896 Fix array indexing of scalar Distribution bug fix Quandela/Perceval 802 Make BSDistribution(bsd) copy the source distribution bug fix Quandela/Perceval 803 Validate each iteration of a ComputationIterator bug fix Quandela/Perceval 804 Set ERROR status when execute_sync fails with allow_partial_results bug fix Quandela/Perceval 805 Reset the Processor source when its noise model is changed bug fix Quandela/Perceval 806 Fix result insertion for a ComputationIterator without iteration bug fix Quandela/Perceval 807 Fix serialization of detectors without a name bug fix Quandela/Perceval 808 Keep min_detected_photons_filter(0) through experiment serialization bug fix Quandela/Perceval 809 Do not copy the experiment in use_phase_noise when there is no phase noise speedup Quandela/Perceval 810 Compute numeric circuit unitaries by updating row blocks speedup scikit-bio/scikit-bio 2676 PERF: slice positions from stacked bytes in TabularMSA.iter_positions speedup lightkurve/lightkurve 1595 Fixed local FITS time units for reordered columns bug fix biotite-dev/biotite 1037 Fix `check_in_box()` returning atoms inside the box bug fix scikit-bio/scikit-bio 2677 BUG: accept a sparse basis in ilr and ilr_inv bug fix lightkurve/lightkurve 1596 Fixed stitching light curves with different time formats bug fix biotite-dev/biotite 1038 Reject GRO stacks with unequal model lengths bug fix dfm/emcee 588 fix: keep autocorrelation error estimates in original steps bug fix healpy/healpy 1152 Fix mmax handling in harmonic_ud_grade bug fix joshspeagle/dynesty 529 fix iteration count for stopped merged runs bug fix healpy/healpy 1153 Fix string column names in write_map bug fix joshspeagle/dynesty 530 fix information for zero reweighting factors bug fix pymc-devs/pymc 8643 Fix MultiTrace slicing counts and negative strides bug fix lmfit/lmfit-py 1068 FIX: use component prefixes in eval_uncertainty bug fix pybamm-team/PyBaMM 6070 fix: preserve skip_ok on repeated experiment steps bug fix phonopy/phonopy 1118 Fix empty arguments in deprecated option name handling bug fix pymc-devs/pymc 8644 Initialize variational groups in their variable order bug fix lmfit/lmfit-py 1069 FIX: handle sequence inputs with nan_policy='omit' bug fix biotite-dev/biotite 1039 Fix GFF annotation grouping for discontinuous features bug fix biotite-dev/biotite 1040 Fix `to_sequence()` chain selection for atom array stacks bug fix scikit-bio/scikit-bio 2678 BUG: drop gap columns across all sequences in AlignPath.to_indices bug fix scikit-bio/scikit-bio 2679 BUG: update stored IDs in OrdinationResults.rename bug fix scikit-bio/scikit-bio 2680 BUG: treat missing lengths as zero while climbing in root_at_midpoint bug fix healpy/healpy 1154 Fix ud_grade averaging with non-finite pixels bug fix joshspeagle/dynesty 531 fix strand IDs when merging independent runs bug fix pymc-devs/pymc 8645 Preserve dimension names in ZarrTrace without coordinate labels bug fix pymc-devs/pymc 8646 Flatten start_sigma arrays when initializing ADVI bug fix lmfit/lmfit-py 1070 FIX: preserve directories when loading saved models from Paths bug fix phonopy/phonopy 1119 Fix loading YAML with a NumPy primitive matrix bug fix phonopy/phonopy 1120 Preserve symbol distinction when replicating Phonopy bug fix phonopy/phonopy 1121 Speed up thermal displacements by batching mode populations speedup pybamm-team/PyBaMM 6072 fix: keep EIS input ordering stable across solves bug fix pybamm-team/PyBaMM 6073 fix: search experiment tags with string steps bug fix exoplanet-dev/celerite2 208 Include GP means in PyMC and NumPyro samples bug fix exoplanet-dev/celerite2 209 Return negative infinity after quiet factorization failures bug fix exoplanet-dev/celerite2 210 Handle PyMC marginal sample sizes bug fix exoplanet-dev/celerite2 211 Reallocate non-owning workspaces after pickling bug fix spglib/spglib 650 fix: preserve legacy Python error messages bug fix spglib/spglib 651 fix: identify magnetic groups in oblique primitive bases bug fix spglib/spglib 652 fix(fortran): forward stabilized reciprocal mesh arguments bug fix spglib/spglib 653 perf: prune lattice symmetry candidates by axis length speedup MultiQC/MultiQC 3738 Fix grouped sample values in get_general_stats_data bug fix MultiQC/MultiQC 3739 Fix module and data-key selection in get_module_data bug fix MultiQC/MultiQC 3740 Preserve section order when merging parquet reports bug fix openmc-dev/openmc 4240 Reshape each mesh filter along its expanded axes bug fix openmc-dev/openmc 4241 Read matrix rotations from mesh filters through the C API bug fix openmc-dev/openmc 4242 Scale shared mesh sources once when setting total strength bug fix openmc-dev/openmc 4243 Preserve mesh identity and transformations when comparing filters bug fix LSSTDESC/CCL 1329 Reject negative neutrino masses bug fix LSSTDESC/CCL 1330 Fix comoving volume in curved cosmologies bug fix LSSTDESC/CCL 1331 Include fc_bar in Duffy08 equality and representation bug fix LSSTDESC/CCL 1332 Fix fourth-profile bias in the two-halo trispectrum bug fix pysam-developers/pysam 1431 Allow VariantRecord POS=0 (start=-1), as used for telomeres bug fix dfm/corner.py 355 Fix linestyle aliases in overplot_points bug fix dfm/corner.py 356 Restore variable selection with ArviZ 1.x bug fix dfm/corner.py 357 Align divergence markers with selected coordinates bug fix mdtraj/mdtraj 2236 Resolve mmCIF connections using label identifiers bug fix mdtraj/mdtraj 2237 Apply atom selection after reading ARC topology bug fix mdtraj/mdtraj 2238 Continue default XTC and TRR frame numbers across writes bug fix mdtraj/mdtraj 2239 Restore integral bond orders when reading HDF5 topology bug fix mdtraj/mdtraj 2240 Use written atom serials in PDB CONECT records bug fix adrn/gala 680 Preserve mock stream metadata through HDF5 bug fix adrn/gala 681 Honor the median option for Staeckel focal lengths bug fix adrn/gala 682 Use elapsed time in fast Lyapunov estimates bug fix adrn/gala 683 Use the toy potential time unit when fitting O2GF frequencies bug fix tenpy/tenpy 779 Fix MPO expectation values with supplied environments bug fix tenpy/tenpy 780 Fix joint probabilities in MPS.sample_measurements bug fix tenpy/tenpy 781 Fix dropping charges on grouped sites of different dimensions bug fix tenpy/tenpy 782 Fix MPO construction for empty Hamiltonians bug fix tenpy/tenpy 783 Preserve Jordan-Wigner parity for grouped independent charges bug fix pyro-ppl/numpyro 2440 fix(infer): reject scanned sample sites in TraceEnum_ELBO bug fix pyro-ppl/numpyro 2441 fix(distributions): keep two-sided TruncatedNormal samples within bounds bug fix pyro-ppl/numpyro 2442 fix(infer): preserve PRNG implementations in Predictive bug fix pyranges/pyranges 413 Preserve strandedness after a DataFrame round trip bug fix pyranges/pyranges 414 Respect as_df and nrows when reading GTF with full=False bug fix tskit-dev/tskit 3510 Stabilize trait correlation across trait scales bug fix tskit-dev/tskit 3511 Preserve edge and migration metadata when trimming bug fix tskit-dev/tskit 3512 Use the converted site mask when validating VCF positions bug fix nipy/nibabel 1637 BF: Use output position when slicing frames of an ECAT array proxy bug fix nipy/nibabel 1638 BF: Read AFNI complex sub-bricks as complex64 bug fix nipy/nibabel 1639 BF: Reverse slice timing when reorientation flips the slice axis bug fix nipy/nibabel 1640 ENH: Write TCK streamlines in blocks when no affine needs applying speedup mdtraj/mdtraj 2241 Build and reduce contact distance groups in bulk speedup adrn/gala 684 Use matrix products for O2GF mode sums speedup tskit-dev/tskit 3513 Avoid per-site copies when writing VCF speedup deeptools/deepTools 1502 fix: keep computeMatrix groups distinct when filenames match bug fix deeptools/deepTools 1503 fix: apply zero thresholds in computeMatrix bug fix deeptools/deepTools 1504 fix: score singleton and two-row heatmap clusters correctly bug fix deeptools/deepTools 1505 fix: preserve missing signal when clustering heatmaps bug fix deeptools/deepTools 1506 fix: write final zero-valued bigwigCompare intervals bug fix open2c/cooler 544 Apply divisive weights in cooler dump bug fix open2c/cooler 545 Write the dump header for empty queries bug fix open2c/cooler 546 Decode contig names in joined pixels bug fix daler/pybedtools 434 preserve input rows when indexing presorted files bug fix daler/pybedtools 435 fix generator completion in randomstats bug fix daler/pybedtools 436 accept pathlib paths in intersection database lists bug fix daler/pybedtools 437 preserve mate position 1 when iterating over BAM bug fix tskit-dev/msprime 2554 Record SLiM mutation model parameters before simulation bug fix tskit-dev/msprime 2555 Preserve initial tables when adding populations bug fix tskit-dev/msprime 2556 Preserve population IDs in ancestry provenance bug fix levitsky/pyteomics 238 Write MGF retention-time ranges bug fix levitsky/pyteomics 239 Apply fixed ProForma modifications at the specified terminus bug fix levitsky/pyteomics 240 Check spectrum IDs in MzMLb membership tests bug fix levitsky/pyteomics 241 Use the charge state passed to ProForma.composition bug fix levitsky/pyteomics 242 Keep PSM identities when streaming FDR-filtered results bug fix California-Planet-Search/radvel 464 Preserve the source basis when converting parameters bug fix California-Planet-Search/radvel 465 Use configured planet letters in dict-based fits bug fix California-Planet-Search/radvel 466 Keep the optimizer's returned fit parameters bug fix California-Planet-Search/radvel 467 Honor fixed jitter in information-criterion comparisons bug fix California-Planet-Search/radvel 468 Sort celerite observation times before initializing the kernel bug fix tskit-dev/tskit 3514 Apply NEXUS node labels to alignment rows bug fix nipy/nibabel 1641 BF: Accept MGH files with an intent code in the version field bug fix nipy/nibabel 1642 BF: Align annotation names with colour table rows when entries are missing bug fix nipy/nibabel 1643 BF: Reshape GIFTI MatrixData to 4x4 when values are on one line bug fix open2c/cooler 547 Read the row index for the queried matrix range speedup daler/pybedtools 438 count BAM and CRAM records without converting intervals speedup deeptools/deepTools 1507 perf: average bigWig coverage bins in NumPy speedup levitsky/pyteomics 243 Reuse log-factorials in q-value correction speedup