10x Genomics
Chromium Single Cell ATAC

Cell Ranger ATAC2.1, printed on 11/17/2024

Cell Ranger ATAC count: Secondary Analysis Results

The cellranger-atac count pipeline outputs several CSV files which contain automated secondary analysis results. A subset of these results are used to render the Cell Clustering view in the run summary.

Dimensionality reduction

Before clustering the cells, Latent Semantic Analysis (LSA) is run on the normalized filtered peak-barcode matrix to reduce the number of feature (peak) dimensions. This produces a projection of each cell onto the first N components (default N=15). One may alternatively choose Principal Component Analysis (PCA) or Probabilistic Latent Semantic Analysis (PLSA) to perform dimensionality reduction in the pipeline. All of these methods provide the following basic CSV output files. Note that the PLSA algorithm in 2.0 is restricted to one thread due to technical reasons and computational performance of dimensionality reduction is likely to be affected.

$ cd /home/jdoe/runs/sample345/outs
$ head -2 analysis/lsa/15_components/projection.csv
Barcode,PC-1,PC-2,PC-3,PC-4,PC-5,PC-6,PC-7,PC-8,PC-9,PC-10,PC-11,PC-12,PC-13,PC-14,PC-15
AAATGAGCAATCAGGG-1,-2.0256188855237585,-20.464971914743963,2.4066208658862194,-0.9789882112497361,-0.09345960806751374,-8.483300343102174,-5.672765504454421,18.312955842984056,5.6927438340737195,-3.0378744705134992,0.3959335790734238,-4.93326991505897,9.485264727952154,0.2107363858043646,0.948135821430962

A components matrix is produced which indicates how much each peak contributed to each component.

$ head -2 analysis/lsa/15_components/components.csv
PC,chr1:9695143-9697582,chr1:9698212-9701041,...
1,-0.5482991923678618,-0.6374211593177428,...

The third file contains the peaks with the highest dispersion that were selected for use in the principal component calculations.

$ head -5 analysis/lsa/15_components/features_selected.csv
Feature
1,chr21:42879060-42879960
2,chr21:44817476-44818342
3,chr21:46323900-46324631
4,chr21:37072809-37073658

The fourth file records the proportion of total variance explained by each principal component. When choosing the number of components that are significant, it is useful to look at the plot of variance explained as a function of component rank - when the numbers start to flatten out, subsequent components are unlikely to represent meaningful variation in the data.

$ head -5 analysis/lsa/15_components/variance.csv
PC,Proportion.Variance.Explained
1,0.8452609977911579
2,0.032765042936590785
3,0.026127180307558735
4,0.01667142686188944

The final file computed by Cell Ranger ATAC lists the normalized dispersion of each peak, after binning peaks by their mean expression across the dataset. This provides a useful measure of variability of each peak.

$ head -5 analysis/lsa/15_components/dispersion.csv
Feature,Normalized.Dispersion
chr1:9695143-9697582,0.02029960904777695
chr1:9698212-9701041,0.10379770925583033
chr1:9825253-9827762,-1.0
chr1:9829746-9830116,25.528012093307737

Visualization

After running dimensionality reduction, t-distributed Stochastic Neighbor Embedding (t-SNE) and Uniform Manifold Approximation and Projection (UMAP) are run to visualize cells in a 2-D space.

$ head -5 analysis/tsne/2_components/projection.csv
Barcode,TSNE-1,TSNE-2
AAATGAGCAATCAGGG-1,1.552159628302055,4.434829693735686
AACAAAGCACCTATTT-1,3.2188609791527667,0.03569781940043248
AACCTTTCAATGATGA-1,-3.8319704788291475,-1.092848944953291
AACTTGGCATGGCCGT-1,-4.226692514189564,0.3351938808086092

$ head -5 analysis/umap/2_components/projection.csv
Barcode,UMAP-1,UMAP-2
AAACTCGTCCAGTTAG-1,2.1193044,2.5859628
AAATGAGCAATCAGGG-1,0.37983754,5.982194
AACAAAGCACCTATTT-1,-1.2387826,5.6922545
AACCGATCAGCAACCC-1,0.16549169,6.803529

Clustering

Clustering is then run to group cells together that have similar accessibility profiles, based on their projection into lower dimensional space. Graph-based clustering (under graphclust) is run once as it does not require prespecification of the number of clusters. For PCA, K-means (under kmeans) is run for many values of K=2,...,N where K corresponds to the number of clusters. For LSA or PLSA, spherical k-means (under kmeans) is run over the same range of K. By default N=5. The corresponding results for each K is separated into its own directory.

$ ls analysis/clustering
graphclust  kmeans_2_clusters  kmeans_3_clusters  kmeans_4_clusters  kmeans_5_clusters

For each clustering, cellranger-atac produces cluster assignments for each cell.

$ head -5 analysis/clustering/kmeans_3_clusters/clusters.csv
Barcode,Cluster
AAATGAGCAATCAGGG-1,2
AACAAAGCACCTATTT-1,1
AACCTTTCAATGATGA-1,3
AACTTGGCATGGCCGT-1,3

Differential enrichment analysis

Prior to differential analysis, cellranger-atac produces a peak-barcode matrix and a transcription factor-barcode matrix of counts as described in Matrices. cellranger-atac then produces a table indicating which peaks and transcription factor motifs are differentially accessible in each cluster relative to all other clusters, as per the algorithms described here. For each feature, whether it is peak or transcription factor motif, we compute three values per cluster:

• The mean cut site counts per cell pooled in feature in cluster i.
• The log2 fold-change of this feature in cluster i relative to other clusters.
• The p-value denoting significance of this feature in cluster i relative to other clusters, adjusted to account for the number of hypotheses for the feature type being tested. So p-value for peaks are adjusted based on the number of peaks being tested, while p-values for transcription factor motifs are adjusted based on the number of transcription factor motifs.

This is located in a different directory than the clustering results, but follows the same structure, with each clustering separated into its own directory.

$ head -5 analysis/enrichment/kmeans_3_clusters/differential_expression.csv
Feature ID,Feature Name,Cluster 1 Mean Counts,Cluster 1 Log2 fold change,Cluster 1 Adjusted p value,Cluster 2 Mean Counts,Cluster 2 Log2 fold change,Cluster 2 Adjusted p value,Cluster 3 Mean Counts,Cluster 3 Log2 fold change,Cluster 3 Adjusted p value
chr1:9695129-9697582,chr1:9695129-9697582,0.014098403818774368,-5.823451487250574,2.2659671842098193e-06,4.185745651762137e-09,-1.3874516676069444,0.5918812904596457,1.9512762483589925,7.238430090771634,5.00258305609651e-09
chr1:9698210-9701041,chr1:9698210-9701041,0.013761153212430422,-6.1502095503083165,7.855686702156565e-07,0.046489553517204636,-3.0232327143356246,0.01647646310191049,2.2844378973176838,6.5025499776936115,4.703658999567952e-13
.
.
.
AHR_HUMAN.H11MO.0.B,AHR_HUMAN.H11MO.0.B,1.5229979744677225e-09,-0.558490289359965,1.0,1.5229979744575502e-09,1.41990325445066,1.0,1.5229979744838465e-09,2.5
360529002402097,1.0
AIRE_HUMAN.H11MO.0.C,AIRE_HUMAN.H11MO.0.C,382.4895824324451,-1.366896997726535,0.007214824200990991,4098.191143669588,0.031632664734601475,1.0,124.229272550
17468,2.136369782757689,0.0015585067057439586

Notice that the table for any specific clustering includes differential analysis results for both peaks and transcription factor motifs.