Cell Ranger ATAC2.0, printed on 11/22/2024
The cellranger-atac reanalyze command reruns secondary analysis performed on the peak-barcode matrix (dimensionality reduction, clustering and visualization) using different parameter settings.
These are the required command line arguments:
| Argument | Description |
|---|---|
--id=ID | A unique run id and output folder name [a-zA-Z0-9_-]+ of maximum length 64 characters. |
--peaks=BED | Specify peaks to use in downstream analyses from supplied 3-column BED file. The
supplied peaks file must be sorted by position and not contain overlapping peaks;
comment lines beginning with # are allowed. |
--fragments=TSV.GZ | Path to the fragments.tsv.gz generated by cellranger-atac count or aggr. Note it is
assumed that the tabix index file fragments.tsv.gz.tbi is present in the same directory. |
--reference=PATH | Path to folder containing a Cell Ranger ATAC or Cell Ranger ARC reference. |
Optional command line parameters are listed below (also available through cellranger-atac reanalyze --help):
| Option | Description |
|---|---|
--description=TEXT | Sample description to embed in output files [default: ]. |
--params=PARAMS_CSV | A CSV file specifying analysis parameters. |
--force-cells=NUM | Define the top N barcodes with the most fragments overlapping peaks as cells. N must be a positive integer <= 20,000. Please consult the documentation before using this option. |
--barcodes=CSV | Specify barcodes to use in analysis. The barcodes could be specified in a text file that contains one barcode per line (blank lines are ignored). Or supply a CSV (with/without a header) whose first column will be used - exports from Loupe Browser will have this format. |
--agg=AGGREGATION_CSV | If the input matrix was produced by cellranger-atac aggr, it is possible to pass the
same aggregation CSV in order to retain per-library tag information in the resulting
.cloupe file. |
--jobmode=MODE | Job manager to use. Valid options: local (default), sge, lsf, slurm or path to a .template file. Search for help on "Cluster Mode" at support.10xgenomics.com for more details on configuring the pipeline to use a compute cluster [default: local]. |
--localcores=NUM | Set max cores the pipeline may request at one time. Only applies to local jobs. |
--localmem=NUM | Set max GB the pipeline may request at one time. Only applies to local jobs. |
--localvmem=NUM | Set max virtual address space in GB for the pipeline. Only applies to local jobs. |
--mempercore=NUM | Reserve enough threads for each job to ensure enough memory will be available, assuming each core on your cluster has at least this much memory available. Only applies to cluster jobmodes. |
--maxjobs=NUM | Set max jobs submitted to cluster at one time. Only applies to cluster jobmodes. |
--jobinterval=NUM | Set delay between submitting jobs to cluster, in ms. Only applies to cluster jobmodes. |
--overrides=PATH | The path to a JSON file that specifies stage-level overrides for cores and memory. Finer-grained than --localcores, --mempercore and --localmem. |
--uiport=PORT | Serve web UI at (http://localhost:PORT). |
After determining input arguments and options, run cellranger-atac
reanalyze. This example reanalyzes the results of an aggregation named
AGG123:
$ cd /home/jdoe/runs
$ ls -1 AGG123/outs/*.gz # verify the input file exists
AGG123/outs/fragments.tsv.gz
$ cellranger-atac reanalyze --id=AGG123_reanalysis \
--peaks=AGG123/outs/peaks.bed \
--params=AGG123_reanalysis.csv \
--reference=/home/jdoe/refs/hg19 \
--fragments=/home/jdoe/runs/AGG123/outs/fragments.tsv.gz
The pipeline will begin to run, creating a new folder named with the reanalysis
ID specified with the --id argument (e.g.
/home/jdoe/runs/AGG123_reanalysis) for its output. If this output
folder already exists, cellranger-atac will assume it is an
existing pipestance and attempt to resume running it.
A successful run should conclude with a message similar to this:
... 2021-04-27 10:38:34 [runtime] (join_complete) ID.AGG123_reanalysis.SC_ATAC_REANALYZER_CS.SC_ATAC_REANALYZER.ATAC_CLOUPE_PREPROCESS Outputs: - Summary of all data metrics: /home/jdoe/runs/AGG123_reanalysis/outs/summary.json - Per-barcode fragment counts & metrics: /home/jdoe/runs/AGG123_reanalysis/outs/singlecell.csv - Raw peak barcode matrix in hdf5 format: /home/jdoe/runs/AGG123_reanalysis/outs/raw_peak_bc_matrix.h5 - Raw peak barcode matrix in mex format: /home/jdoe/runs/AGG123_reanalysis/outs/raw_peak_bc_matrix - Filtered peak barcode matrix in hdf5 format: /home/jdoe/runs/AGG123_reanalysis/outs/filtered_peak_bc_matrix.h5 - Filtered peak barcode matrix in mex format: /home/jdoe/runs/AGG123_reanalysis/outs/filtered_peak_bc_matrix - Directory of analysis files: /home/jdoe/runs/AGG123_reanalysis/outs/analysis - HTML file summarizing aggregation analysis : /home/jdoe/runs/AGG123_reanalysis/outs/web_summary.html - Filtered tf barcode matrix in hdf5 format: /home/jdoe/runs/AGG123_reanalysis/outs/filtered_tf_bc_matrix.h5 - Filtered tf barcode matrix in mex format: /home/jdoe/runs/AGG123_reanalysis/outs/filtered_tf_bc_matrix - Loupe Browser input file: /home/jdoe/runs/AGG123_reanalysis/outs/cloupe.cloupe - Annotation of peaks with genes: /home/jdoe/runs/AGG123_reanalysis/outs/peak_annotation.tsv Pipestance completed successfully!
Refer to the Analysis page for an explanation of the output.
The CSV file passed to --params should have one row for every
parameter that you want to customize. There is no header row. If a parameter is
not specified in your CSV, its default value will be used. See Common Use
Cases for some examples.
Here are detailed descriptions of each parameter. For parameters that subset the
data, a default value of null indicates that no subsetting happens
by default.
| Parameter | Type | Default Value | Recommended Range | Description |
|---|---|---|---|---|
dim_reduce | str | lsa | [lsa, pca, plsa] | Pick dimensionality reduction technique. Note: plsa has been temporarily
restricted to run in single-threaded mode due to technical considerations. This could
lead to a longer wall time for execution as compared to v1.2. Multi-threading will be
restored in a subsequent release |
num_analysis_bcs | int | null | Cannot be set higher than the available number of cells or lower than zero. | Randomly subset data to N barcodes for all analyses. Reduce this parameter if you want to improve performance or simulate results from lower cell counts. Resets to available number of cells if specified to be higher than it. |
num_dr_bcs | int | null | Cannot be set higher than the available number of cells. | Randomly subset data to N barcodes when computing PCA projection (the most memory-intensive step). The PCA projection will still be applied to the full dataset, i.e. your final results will still reflect all the data. Try reducing this parameter if your analysis is running out of memory. |
num_dr_features | int | null | Cannot be set higher than the number of peaks in the BED file. | Subset data to the top N features (that is, peaks, ranked by normalized dispersion) when computing LSA/PCA/PLSA projection (the most memory intensive step). The dimreduce projection will still be applied to the full dataset, i.e. your final results will still reflect all the data. Try reducing this parameter if your analysis is running out of memory. |
num_comps | int | 15 | 10-100 (20 for PLSA), depending on the number of cell populations / clusters you expect to see. | Compute N principal components for LSA/PCA/PLSA. Setting this too high may cause spurious clusters to be called. |
graphclust_neighbors | int | 0 | 10-500, depending on desired granularity. | Number of nearest-neighbors to use in the graph-based clustering. Lower values result in higher-granularity clustering. The actual number of neighbors used is the maximum of this value and that determined by neighbor_a and neighor_b. Set this value to zero to use those values instead. |
neighbor_a | float | -230.0 | Determines how clustering granularity scales with cell count. | The number of nearest neighbors, k, used in the graph-based clustering is computed as follows: k = neighbor_a + neighbor_b *log10(n_cells). The actual number of neighbors used is the maximum of this value and graphclust_neighbors. |
neighbor_b | float | 120.0 | Determines how clustering granularity scales with cell count. | The number of nearest neighbors, k, used in the graph-based clustering is computed as follows: k = neighbor_a + neighbor_b* log10(n_cells). The actual number of neighbors used is the maximum of this value and graphclust_neighbors. |
max_clusters | int | 10 | 10-50, depending on the number of cell populations / clusters you expect to see. | Compute K-means clustering using K values of 2 to N. Setting this too high may cause spurious clusters to be called. |
tsne_input_pcs | int | null | Cannot be set higher than the num_comps parameter. | Subset to top N principal components for TSNE. Change this parameter if you want to see how the TSNE plot changes when using fewer PCs, independent of the clustering / differential expression. You may find that TSNE is faster and/or the output looks better when using fewer PCs. |
tsne_perplexity | int | 30 | 30-50 | TSNE perplexity parameter (see the TSNE FAQ for more details). When analyzing 100k+ cells, increasing this parameter may improve TSNE results, but the algorithm will be slower. |
tsne_theta | float | 0.5 | Must be between 0 and 1. | TSNE theta parameter (see the TSNE FAQ for more details). Higher values yield faster, more approximate results (and vice versa). The runtime and memory performance of TSNE will increase dramatically if this is set below 0.25. |
tsne_max_dims | int | 2 | Must be 2 or 3. | Maximum number of TSNE output dimensions. Set this to 3 to produce both 2D and 3D TSNE projections (note: runtime will increase significantly). |
tsne_max_iter | int | 1000 | 1000-10000 | Number of total TSNE iterations. Runtime increases linearly with number of iterations. |
tsne_stop_lying_iter | int | 250 | Cannot be set higher than tsne_max_iter. | Iteration at which TSNE learning rate is reduced. |
tsne_mom_switch_iter | int | 250 | Cannot be set higher than tsne_max_iter. | Iteration at which TSNE momentum is reduced. Cannot be set higher than tsne_max_iter. |
random_seed | int | 0 | any integer | Random seed. Due to the randomized nature of the algorithms, changing this will produce slightly different results. |
These examples illustrate what could be included in the --params
CSV file in some common situations.
For very large or diverse cell populations, the defaults may not capture the full variation between cells. In that case, try increasing the number of principal components and/or clusters. To run dimensionality reduction with 50 components and k-means with up to 30 clusters, include this in the CSV:
num_comps,50 max_clusters,30
You can limit the memory usage of the analysis by computing the LSA projection on a subset of cells and features. This is especially useful for large datasets (100k+ cells). If you have 100k cells, it is reasonable to use only 50% of them for LSA - the memory usage will be cut in half, but you will still be well equipped to detect rare subpopulations. Limiting the number of features will reduce memory even further. To compute the LSA projection using 50000 cells and 3000 peaks, include this in the CSV:
num_dr_bcs,50000 num_dr_features,3000
Note: subsetting of cells is done randomly, to avoid bias. Subsetting of features is done by binning features by their mean expression across cells, then measuring the dispersion (a variance-like parameter) of each gene's expression normalized to the other features in its bin.