v1.7.9

What's Changed

• better skill discovery by @HendricksJudy in #483
• improvements and refactorings across the codebase, focusing on skill packaging and validation, Unicode output support, skill registry robustness, and defensive programming. by @HendricksJudy in #487
• robustness and user-friendliness of the single-cell analysis skills documentation by @HendricksJudy in #488
• Fix: Correct 'X_harmony' KeyError in lazy_function, lazy_checkpoint and by @Aegon8719 in #493
• Fix highly_variable_features mask not updating after HVG selection by @Starlitnightly in #497
• Fix KeyError in ov.pl.embedding when use_raw=True by @Starlitnightly in #498
• Fix ValueError in VIA pruning_clustergraph: correct mask array size by @Starlitnightly in #499
• Fix fq_data_preprocess bug #501: Add paired/single-end data type support and improve tool checks by @zhiluo20 in #505
• Real-world prompt library for the ov.Agent by @HendricksJudy in #502
• Robust filesystem-based context management by @HendricksJudy in #507
• Bug fix and function improvement by @zhiluo20 in #510
• feat(space): add FlashDeconv as third deconvolution method by @cafferychen777 in #519
• Restrict pandas version to <3.0.0 by @DBinary in #529
• OV Agent backend improvements by @HendricksJudy in #530
• Bulk Alignment Optimization and Reconstruction by @zhiluo20 in #531
• Generate multi-sample count matrix CSV instead of per-sample TXT outputs by @zhiluo20 in #533
• fix: handle missing gene group colors in dotplot by @Starlitnightly in #534
• Refactor alignment module: extract CLI utilities and add bulk RNA-seq pipeline by @HendricksJudy in #535
• Integrate BioContext MCP for real-time external database queries by @HendricksJudy in #537
• Revert "Integrate BioContext MCP for real-time external database queries" by @HendricksJudy in #539

New Contributors

• @cafferychen777 made their first contribution in #519

Full Changelog: v1.7.8...v1.7.9

v1.7.8

v 1.7.8

Implemented lazy loading system that reduces import omicverse time by 40% (from ~7.8s to ~4.7s).
Added GPU-accelerated PCA support for Apple Silicon (MLX) and CUDA (TorchDR) devices.
Introduced Smart Agent System with natural language processing for 50+ AI models from 8 providers.
Added and fixed the anndata-rs to support million size's datasets (#336)

PP Module

• Added GPU-accelerated PCA in ov.pp.pca() with MLX support for Apple Silicon MPS devices
• Added TorchDR-based PCA acceleration in ov.pp.pca() for NVIDIA CUDA devices
• Added smart device detection and automatic backend selection in init_pca() and pca() functions
• Added graceful fallback to CPU implementation when GPU acceleration fails
• Added enhanced verbose output with device selection information and emoji indicators
• Added optimal component determination based on variance contribution thresholds in init_pca()
• Added GPU-accelerated SUDE dimensionality reduction in ov.pp.sude() with MLX/CUDA support
• Optimize the ov.pp.qc and added ribosome and hb-genes to know more information of data quantity.

Datasets Module

• Complete elimination of scanpy dependencies for faster loading
• Added dynamo-style dataset framework with comprehensive collection
• Added robust download system with progress tracking and caching
• Added enhanced mock data generation with realistic structure
• Added support for h5ad, loom, xlsx, and compressed formats

Agent Module

• Added multi-provider LLM support (OpenAI, Anthropic, Google, DeepSeek, Qwen, Moonshot, Grok, Zhipu AI)
• Added natural language processing for both English and Chinese
• Added code generation architecture with local execution
• Added function registry system with multi-language aliases
• Added smart API key management and provider-specific configuration

Bulk Module

• Added BayesPrime and Scaden to deconvoluted Bulk RNA-seq's celltype proportion.
• Added alignment to alignment the fastq to counts.

Single Module

• Added ov.single.Annotation and ov.single.AnnotationRef to annotate the cell type automatically.
• Added ov.alignment.single to alignment the scRNA-seq to counts directly.

v1.7.6

LLM Module

• Added GeneFromer, scGPT, scFoundation, UCE, CellPLM to call directly in OmicVerse.

Pl Module

• Optimized the visualization effect of embedding.
• Added ov.pl.umap, ov.pl.pca, ov.pl.mde, and ov.pl.tsne

v1.7.2

Pl Module

• Optimated the plot effect of ov.pl.box_plot
• Optimated the plot effect of ov.pl.volcano
  Optimated the plot effect of ov.pl.violin
• Added beautiful dotplot than scanpy (#318)
• Added the similar visualization function of CellChat. (#313)

Space Module

• Added 3D cell-cell interaction analysis in COMMOT (#315)

Single Module

• Fixed the error of pathway_enrichment. (#184)
• Added SCENIC module with GPU-accerlate. (#331)

utils Module

• Added scICE to calculate the best cluster (#329)

v1.7.1

Single Module

• Fixed some error of ov.single.lazy.
• Fixed the format of ov.single.generate_scRNA_report
• Updated some functions of palantir
• Added CellOntologyMapper to map cell name.

CellOntologyMapper is a very useful function. Hopefully it would be help for your job.

Tutorial of CellOntologyMapper: https://omicverse.readthedocs.io/en/latest/Tutorials-single/t_cellmatch/

v1.7.0

Added cpu-gpu-mixed to accelerate the analysis of scrna-seq using GPU.
Changed the logo presentation of Omicverse to ov.plot_set

Bulk Module

• Added limma, edgeR in different expression gene analysis. (#238)
• Fixed the version error of DEseq2 analysis.

Single Module

• Added lazy function to calculate all function of scrna-seq (#291)
• Added generate_scRNA_report and generate_reference_table to generate the report and reference (#291) (#292)
• Fixed geneset_prepare not being able to read gmt not split by \t\t (#235) (#238)
• Added geneset_aucell_tmp,pathway_aucell_tmp,pathway_aucell_enrichment_tmp to test the chunk_size (#238)
• Added data enhancement of Fate
• Added plot_atlas_view_ov in VIA
• Fixed an error when the matrix is too large in recover_counts.
• Added forceatlas2 to calculate the X_force_directed.
• Added milo and scCODA to analysis different celltype abundance.
• Added memento to analysis different gene expression.

Space Module

• Added GASTON to learn a topographic map of a tissue slice from spatially resolved transcriptomics (SRT) data (#238)
• Added super kwargs in plot_tensor_single of STT.
• Updated COMMOT using GPU-accerlate

Plot Module

• Added dotplot_doublegroup to visual the genes in doublegroup.
• Added transpose argument of cpdb_interacting_heatmap to transpose the figure.
• Added calculate_gene_density to plot the gene's density.

v1.6.9

v 1.6.9

PP Module

• Added recover_counts to recover counts after ov.pp.preprocess
• removed the lognorm layers added in ov.pp.pca

Single Module

• Added MultiMap module to integrate multi species
• Added CellVote to vote the best cells
• Added CellANOVA to integrate samples and correct the batch effect
• Added StaVia to calculate the pseudotime and infer trajectory.

Space Module

• Added ov.space.cluster to identify the spatial domain
• Added Binary for spatial cluster
• Added Spateo to calculate the SVG

v1.6.8

Bulk Module

• Fixed the error of log_init in gsea_obj.enrichment (#184)
• Added ax argument to visualize the geneset_plot

Space Module

• Added CAST to integrate multi slice
• Added crop_space_visium in omicverse.tl to crop the sub area of space data

Pl Module

• Added legend argument to visualize the cpdb_heatmap
• Added text_show argument to visualize the cellstackarea
• Added ForbiddenCity color system

v1.6.7

Space Module

• Added n_jobs argument to adjust thread in extenel.STT.pl.plot_tensor_single
• Fixed an error in extenel.STT.tl.construct_landscape
• Updated the tutorial of COMMOT and Flowsig

Pl Module

• Added legend_awargs to adjust the legend set in pl.cellstackarea and pl.cellproportion

Single Module

• Fixed the error of get_results and get_results_rfc in cNMF module. (#143) (#139)
• Added sccaf to obtain the best clusters.
• Fixed the .str error in cytotrace2 (#146)

Bulk Module

• Fixed the import error of gseapy in bulk.geneset_enrichment
• Optimized code logic for offline enrichment analysis, added background parameter
• Added pyWGCNA package replace the raw calculation of pyWGCNA (#162)

Bulk2Single Module

• Remove _stat_axis in bulk2single_data_prepare and use index instead of it (#160).

PP Module

• Fixed a return bugs in pp.regress_and_scale (#156)
• Fixed a scanpy version error when using ov.pp.pca (#154)

v1.6.6

Pl Module

• Fixed the 'celltyep_key' error of ov.pl.cpdb_group_heatmap #109
• Fixed an error in ov.utils.roe when some expected frequencies are less than expected value.
• Added cellstackarea to visual the Percent stacked area chart of celltype in samples.

Single Module

• Fixed the bug of ov.single.cytotrace2 when adata.X is not sparse data. #115, #116
• Fixed the groupby error in ov.single.get_obs_value of SEACells.
• Fixed the error of cNMF #107, #85
• Fixed the plot error when Pycomplexheatmap version > 1.7 #136

Bulk Module

• Fixed an key error in ov.bulk.Matrix_ID_mapping
• Added enrichment_multi_concat in ov.bulk to concat the result of enrichment.
• Fixed the pandas version error in gseapy #137

Bulk2Single Module

• Added adata.var_names_make_unique() to avoid mat shape error if gene not unique. #100

Space Module

• Fixed an error in construct_landscape of ov.space.STT
• Fixed an error of get_image_idx_1D in ov.space.svg #117
• Added COMMOT to calculate the cell-cell interaction of spatial RNA-seq.
• Added starfysh to deconvolute spatial transcriptomic without scRNA-seq (#108)

PP Module

• Updated constraint error of ov.pp.mde #129
• Fixed type error of float128 #134