10x Genomics
Chromium Single Cell Gene Expression

Cell Ranger7.1 (latest), printed on 11/21/2024

Release notes for Cell Ranger 7.1.0 (December 7, 2022):

Changes that apply to Fixed RNA Profiling analysis

1. v1.0.1 probe set reference CSVs for human and mouse have a new region column, which indicates whether a probe spans a splice junction by at least 10 bp (spliced) or not (unspliced).
2. When the v1.0.1 probe set reference CSV is used in Cell Ranger v7.1, the web summary and metrics_summary.csv files will include genomic DNA metrics. The region column information is used to calculate these metrics.
3. The molecule_info.h5 files include the probe to which each molecule is mapped.
4. In the probe set reference CSV, the included column is set toFALSE for all deprecated probes.

General improvements

1. Calling cell barcode improvement: The auto-estimated expect-cells upper range has been restricted to the lower range of the EmptyDrops method. Previously, the upper range was 262,000 cells. The new upper range is 45,000 for single cell gene expression analyses. The exception is super-loaded multiplex Fixed RNA Profiling analyses, where the range is calculated as: max(45,000, number of probe barcodes × 22,500).
2. Improvement to the Batch Effect Score (BES) calculation to use √n nearest neighbors, where n is the total number of cells, instead of 100. Cells are no longer subsampled to 10%.
3. A new compression format, .tar.xz, is available on the Cell Ranger downloads page. The smaller file size enables faster download.
4. Cell Ranger 7.1 introduces a new subcommand, cellranger multi-template, which provides descriptions for all multi config CSV parameters and produces a config CSV template. Run cellranger multi-template -h for help.
5. The ARC-v1 chemistry may be used to analyze only the Gene Expression library portion of a Multiome ATAC + Gene Expression experiment.
6. The aggregate_barcodes.csv output file for Antibody Capture analyses is no longer stored in a antibody_analysis/ sub-directory. In the cellranger multi pipeline, it is found in outs/per_sample_outs/<sample_name>/count/aggregate_barcodes.csv. In the cellranger count pipeline, it is found in outs/aggregate_barcodes.csv. For both pipelines, the CSV file is only generated if antibody aggregates are detected.
7. The web summaries for Antibody Capture libraries include a Distribution of Antibody Counts plot to show the relative composition of antibody counts for antibodies with at least one UMI.

Bug fixes

1. Fixed a bug in the OrdMag algorithm, which could result in all barcodes being called as cells when there are very few cells.
2. Improvement to 3' Cell Multiplexing tag assignment for samples with a large number of zero CMO UMI counts.
3. Fixed a 3' Cell Multiplexing t-SNE plot bug where the plots were generated assuming a full set of CMOs in cmo-set instead of those used in the experiment.
4. Fixed conditions resulting in negative gap errors.

New feature: Barcode Enabled Antigen Mapping (BEAM) or Antigen Capture

1. Cell Ranger 7.1 is required for the analysis of BEAM libraries. Instructions for running cellranger multi are described in the Antigen Capture page. The new feature_type, antigen capture, the Feature Reference CSV that specifies the list of antigens (and MHC alleles) included in the experiment, and all the antigen specific customizable parameters in a multi config CSV are described in detail. Example multi config CSV for TCR and BCR Antigen Capture libraries are also provided.
2. The algorithms section includes a page called Antigen Algorithms with a description of the new methods developed for processing Antigen Capture (BEAM) data.
3. If an Antigen Capture library is included, some new/updated output files are generated (described in the Understanding Outputs section):
   • antigen_specificity_scores.csv (new file)
   • per_barcode.csv (new file)
   • aggregate_barcodes.csv (updated location and format)

Changes that apply to 5' Immune Profiling analysis

1. V(D)J cell calling improvement: If a Gene Expression library is present, the V(D)J cell calling algorithm does not filter out two or more clonotypes that have identical chains. This helps improve V(D)J cell calling, especially for transgenic strains. This change does not apply to V(D)J datasets in the absence of a Gene Expression library.
2. The Human V(D)J reference has been updated to exclude the following genes:

• IGHV4-30-2
• IGKV1D-33
• IGKV1D-37
• IGKV1D-39
• IGKV2D-28

These genes have counterparts with identical V, D, J, and C gene sequences, but differ in the length of their 5' UTRs. Removing duplicates improves clonotype assignment.

Release notes for Cell Ranger 7.0.1 (August 18, 2022):

Bug fixes

1. Fixed a bug where an upgrade to Illumina NovaSeq control software v1.8 (reagent name change in recipe XML file) resulted in a silent cellranger mkfastq error and a significant number of reads going into Undetermined/ because the orientation of i5 (Index2) could not be autodetected.
2. Improved Deplex Error message for cellranger multi when no valid cell multiplexing tags are detected in the Multiplexing Capture library. Common failure modes are provided to help with troubleshooting.
3. Updated Fixed RNA Profiling web summary metric names and definitions for consistency.

Release notes for Cell Ranger 7.0.0 (May 17, 2022):

New feature: Fixed RNA Profiling

1. Cell Ranger 7.0 is required for analysis of Fixed RNA Profiling data. Instructions for running the cellranger multi subcommand are described in the running multi pipeline page. This includes a new option, probe-set, to specify the probe set CSV file. Output files are described in the Understanding Outputs section. The Fixed RNA Profiling algorithms section includes descriptions of the new methods that were developed for processing Fixed RNA Profiling data.

Major updates

1. To maximize sensitivity for whole transcriptome 3’/5’ Single Cell Gene Expression and 3’ Cell Multiplexing experiments, introns will be included in the analysis by default for cellranger count and multi. There will be an informational alert in the count and multi web summaries to indicate that intronic reads were included in your analysis. While not recommended, users can exclude introns by setting include-introns=false in Cell Ranger. This change does not apply to the 3’/5’ Targeted Gene Expression or Fixed RNA Profiling assays, as both target exonic sequences. Learn more here.

2. CRISPR Guide Capture libraries can be aggregated with cellranger aggr. This addition allows users to combine large CRISPR assays across multiple GEM wells. There are no changes to aggr inputs – the presence of CRISPR libraries in the molecule_info.h5 input files enables CRISPR aggregation. Normalization is enabled by default for both Gene Expression and CRISPR libraries; changes to the normalization parameters affect both libraries. Protospacer calling is performed again on the combined data included in the cellranger aggr run. CRISPR aggregation generates the crispr_analysis folder in the outs/ directory. The structure of the crispr_analysis folder is similar to the CRISPR outputs from count.

General improvements

1. Users no longer need to specify expect-cells for cellranger count and multi pipelines due to improvements in the gene expression cell calling algorithm. The expected number of cells can either be auto-estimated (recommended) or users can still provide a reasonable estimate to expect-cells.

2. The new check-library-compatibility option allows users to disable the default check for 10x Barcode overlap when multiple libraries are specified for cellranger count and multi (3' Gene Expression, 5' Immune Profiling).

3. For 3’ Cell Multiplexing analysis in cellranger multi, users can override Cell Ranger’s cell calling and tag calling algorithms with the custom cell assignment input file specified by the barcode-sample-assignment option in the multi config CSV file.

4. Modifications to the 3’ Cell Multiplexing CMO tag calling algorithm enable users to recover viable singlet data from “blank” assignments.

5. The following per-sample output files from cellranger multi have been renamed:

6. Secondary analysis outputs will be named to reflect which library they are specific to (gene_expression_*, antibody_capture_*, crispr_guide_capture_*, multiplexing_capture_*). The secondary analysis clustering, PCA/t-SNE/UMAP, and differential gene expression outputs are supported for Gene Expression and Antibody Capture libraries, while PCA/t-SNE/UMAP outputs are supported for CRISPR Guide Capture and Cell Multiplexing libraries. For example:

└── analysis
    └── pca
        ├── antibody_capture_10_components/
        └── gene_expression_10_components/

7. The cellranger count web summary “Analysis” tab has been renamed to “Gene Expression”. There is an “Antibody” tab for Antibody Capture analysis, which includes a t-SNE projection plot by clustering and a histogram of antibody counts.

8. The cellranger multi web summary (3' Gene Expression, 5' Immune Profiling) “Sample” view has been renamed to “Cells”. The “Antibody” tab includes a t-SNE projection plot by clustering. The mapping metrics, sequencing saturation plot, and median genes per cell plot are displayed on the “Library” view (previously appeared on “Sample” and “Library” view).

9. Cell Ranger can now ingest FASTQs with a quality score up to the full supported range (93 instead of 41).

Bug fixes

1. Improved error messages and better handling of poorly formatted inputs in cellranger mkref. Enable users to generate references for analyses with large genomes containing chromosomes longer than 512 Mbp. cellranger count and multi pipelines will output a .csi BAM index file instead of .bai in these cases.

2. Fixed a bug that resulted in a segmentation fault error when mapping to references that contain small contigs, for example, the rabbit genome.

3. Removed the Inconsistent Throughput Detected alert in web summary when it should not appear.

4. Fixed a bug where VDJ pipeline failed for specific CentOS/RHEL 7 kernels.

5. Bundles the latest version of bamtofastq (v1.4.1) in Cell Ranger 7.0 tarball.

6. Fixed a bug where bamtofastq failed if the R1 read length was >26bp.

Changes that apply to 5' Immune Profiling analysis

1. Support for gamma-delta libraries: The cellranger multi pipeline can process T cell receptor (TCR) libraries enriched for gamma (TRG) and delta (TRD) chains. 10x Genomics does not officially support TRG/D analysis with a reagent kit. Please note that, only CDR3 annotation is available for TRG/D, and the quality of annotations cannot be guaranteed. Users must specify VDJ-T-GD as the feature_type in the cellranger multi config CSV as TRG/D chains cannot be autodetected. A web_summary alert is displayed to indicate the use of an unsupported workflow. No TRG/D analysis is available via the cellranger vdj pipeline.

2. V(D)J Reference updated: The recommended V(D)J reference packages for human and mouse have been updated from version 5.0 to 7.0. The changes to the V(D)J reference sequences are listed below:

HUMAN:

• Added human IGHV3-9
• For two genes that are identical except for extra bases on the 3' end, only the longer version was retained. List of affected genes:

MOUSE:

• Added missing mouse TRGV and TRGC genes

• For two genes that are identical except for extra bases on the 3' end, only the longer version was retained. List of affected genes:

3. V(D)J web summary: In the web_summary.html file produced by cellranger vdj, the Analysis tab has been renamed to VDJ.

4. The default for the fiveprime_multiplexing parameter in the cellranger-7.0.0/lib/bin/parameters.toml file has changed to True.