v1.5.0

R package cellxgene.census V1 is out

The Census team is pleased to announce the release of the R package cellxgene.census.

This has been long coming since our Python release back in May. Now, from R, computational biologists can access the Census data which is the largest standardized aggregation of single-cell data, composed of >30M cells and >60K genes.

With cellxgene.census in a few seconds users can access any slice of Census data using cell or gene filters across hundreds of datasets. The data can be fetched in an iterative fashion for bigger-than-memory slices of data, or quickly exported to basic R structures, and Seurat or SingleCellExperiment for downstream analysis.

Make sure to checkout the R docs from our doc-site, links can be found in the github repo chanzuckerberg/cellxgene-census.

Additions

• [builder] add normalization layer and computed axes statistics by @bkmartinjr in #612
• [builder] add dataset_version_id by @bkmartinjr in #618
• [builder] implement worker count limit by @bkmartinjr in #673
• [python] PCA demo notebook by @bkmartinjr in #580
• [r] Add get_single_cell_experiment by @pablo-gar in #640

Fixes

• [docs] cell guide button beta tag alignment by @ebezzi in #657
• [builder] fix merge error in schema doc by @bkmartinjr in #662
• [docs] Add mean_variance notebook symlink by @ebezzi in #677
• [docs] Fix number of unique cells for human in doc site by @pablo-gar in #678
• [r] restore vignettes to docsite (in static HTML form) by @mlin in #583
• [r] Fix copying R docs, doc-site CI by @pablo-gar in #681

What's Changed

• [builder] bump Census build minimum disk space requirement to 1.8TiB by @bkmartinjr in #661
• [python] bound pytorch memory usage by @atolopko-czi in #654
• [docs] Fetch docsite version number from git tags by @ebezzi in #675
• [docs] fetch-depth: 0 for docsite build by @ebezzi in #679
• [r] Audit r vignettes by @pablo-gar in #686
• [r] Update R docs and doc-site by @pablo-gar in #684
• [r] run R tests with both release and latest versions of tiledbsoma by @mlin in #691
• [r] Change R Dependencies by @pablo-gar in #693
• [r] Update DESCRIPTION to remove windows support from R by @pablo-gar in #692
• Suggesting very minor edits to comp_bio_data_integration.Rmd by @eddelbuettel in #694
• [r] Update to version 1.5.0 and rebuild doc-site by @pablo-gar in #695

New Contributors

• @eddelbuettel made their first contribution in #694

Full Changelog: v1.4.0...v1.5.0

v1.4.0

Additions

• Experimental module for calculating out-of-core mean and variance

Fixes/Changes

• Documentation changes to reflect the new LTS release 2023-07-25

What's Changed

• Bump aiohttp from 3.8.4 to 3.8.5 in /tools/cellxgene_census_builder by @dependabot in #632
• fix: dialog by @tihuan in #633
• Expose out-of-core Mean and Variance APIs by @ebezzi in #609
• add a dataset blocklist by @bkmartinjr in #636
• dataset blocklist location check by @bkmartinjr in #637
• feat(newsletterBanner): Add localStorage by @tihuan in #643
• Switch to prod role for the builder GHA job by @ebezzi in #644
• [r] hotfix for release directory JSON parsing problem by @mlin in #649
• Update builder to use schema_version param to pin dataset CxG schema versions to 3.0.0 by @nayib-jose-gloria in #650
• Updates docs and test results for LTS data release 2023-07-25 by @pablo-gar in #652
• Notebooks for LTS 2023-07-25 by @ebezzi in #656
• docs: add cell guide button by @ebezzi in #655

New Contributors

• @dependabot made their first contribution in #632
• @nayib-jose-gloria made their first contribution in #650

Full Changelog: v1.3.1...v1.4.0

v1.3.1

Fixes/changes:

• Documentation and docsite improvements
• Fix issue #622, inconsistent default census_version used in various API endpoints

What's Changed

• pin builder dependencies by @bkmartinjr in #594
• pytorch eager fetch follow-up by @atolopko-czi in #592
• Adds sparse iter reference docs to doc-site by @pablo-gar in #597
• docs: Add newsletter banner by @tihuan in #605
• change error handling for /curation REST errors by @bkmartinjr in #608
• Add R acceptance tests, update Python acceptance tests by @pablo-gar in #576
• Expand doc-site section "Census data releases" and add data erratum by @pablo-gar in #600
• [r] get_seurat: allow selection of Seurat counts or data slot by @mlin in #607
• fixes minor documentation issues by @pablo-gar in #619
• Complete docstrings for PyTorch ExperimentDataPipe API by @atolopko-czi in #613
• consistently set default census version to stable by @bkmartinjr in #623
• add missing census_version param by @bkmartinjr in #624
• hard error if the builder receives an unsupported CXG schema version by @bkmartinjr in #625
• add do_not_delete flag to release.json format by @bkmartinjr in #628

New Contributors

• @tihuan made their first contribution in #605

Full Changelog: v1.3.0...v1.3.1

v1.3.0

Breaking changes

• Remove Python 3.7 support

Additions

• New sparse iterator added to experimental package (#591)

Fixes / changes

• Pytorch loader performance improvements and bug fixes
• Update dependency pins for multiple packages
• Documentation updates

What's Changed

• fix incorrect pytorch obs soma_joinids by @atolopko-czi in #555
• Fix MacOS failing tests by @ebezzi in #557
• remove pytorch_lr_classifier.py by @atolopko-czi in #556
• Fix doc-site link in README.md by @pablo-gar in #571
• Correct pytorch notebook header by @pablo-gar in #570
• Avoid unnecessary re-initialization of ExperimentDataPipe in mp mode by @atolopko-czi in #566
• census cell dup check by @bkmartinjr in #569
• require s3fs>=2021.06.1 where cache_regions kwarg is added by @RemyLau in #572
• Notebook reformatting by @ebezzi in #579
• [r] install tiledbsoma from github by @mlin in #589
• Remove python 3.7 support by @bkmartinjr in #582
• use less mem for hvg tests by @bkmartinjr in #590
• Add eager fetching of SOMA batches to ExperimentDataPipe by @atolopko-czi in #575
• add sparse iterator to experimental.utils by @bkmartinjr in #591

New Contributors

• @RemyLau made their first contribution in #572

Full Changelog: v1.2.1...v1.3.0

v1.2.1

Additions

• Python 3.11 support
• add notebooks to demonstrate new experimental APIs

Fixes

• Pytorch loader multi-process issues [#522, #516]
• Doc site improvements

What's Changed

• fix: pytorch unit test hangs by @atolopko-czi in #522
• Add support for Python 3.11 by @martinkim0 in #528
• update notebook guidelines by @pablo-gar in #534
• Add docs for experimental modules by @ebezzi in #537
• demo notebook for HVG experimental API by @bkmartinjr in #536
• Add experimental notebooks to docsite + fix headers by @ebezzi in #541
• fix OOM on pytorch unit test by @atolopko-czi in #542
• Add pytorch notebook by @atolopko-czi in #551

New Contributors

• @martinkim0 made their first contribution in #528

Full Changelog: v1.2.0...v1.2.1

v1.2.0

Add experimental API cellxgene_census.experimental.pp.get_highly_variable_genes

What's Changed

• minor clarifications for the pypi.org release process by @atolopko-czi in #512
• [r] cache most R dependencies to speed up r-check CI by @mlin in #517
• [r] Add comp_bio_census_info.Rmd by @mlin in #407
• fix pytorch multiprocessing result by @atolopko-czi in #516
• Update release_process.md by @atolopko-czi in #520
• highly variable gene annotation by @bkmartinjr in #511

Full Changelog: v1.1.0...v1.2.0

v1.1.0

tiledbsoma upgraded to 1.2.5. Added an experimental PyTorch data loader.

What's Changed

• [docs] Add autosummary by @ebezzi in #492
• Add Databricks install instructions to FAQ by @atolopko-czi in #488
• [docs] Fix the Census link in navbar by @ebezzi in #491
• PyTorch DataLoader by @atolopko-czi in #499
• chore: update lifecycle tags by @atolopko-czi in #509
• bump python api tiledbsoma version by @atolopko-czi in #510

Full Changelog: v1.0.2...v1.1.0

v1.0.2

What's Changed

Update to tiledbsoma=1.2.4, which adds support for MacOS ARM64.

Change log summary:

• Bump static census version in R tests by @ebezzi in #472
• [r] update get_presence_matrix() and vignette to use zero-based matrix view by @mlin in #475
• add acceptance test run by @atolopko-czi in #485
• rerun notebooks for census build 2023-05-15 by @atolopko-czi in #484
• Updated gget cellxgene tutorial to reflect workflow updates [formatting corrected] by @pablo-gar in #490
• Add docsite version number from the library version by @ebezzi in #481
• pin tiledbsoma==1.2.4 by @bkmartinjr in #493

Full Changelog: v1.0.1...v1.0.2

v1.0.1

What's Changed

• Enable anonymous access to S3 bucket by @ebezzi in #275

Full Changelog: v1.0.0...v1.0.1

v1.0.0rc4

What's Changed

• revert R API release.json URL by @atolopko-czi in #458
• Fix docsite version and disable searchbar by @ebezzi in #460
• update release bootstrap URL to public permalink by @bkmartinjr in #459

Full Changelog: v1.0.0rc3...v1.0.0rc4