In bioinfocz/scdrake: A pipeline for droplet-based single-cell RNA-seq data secondary analysis implemented in the drake Make-like toolkit for R language

Louvain algorithm {.tabset}

Louvain algorithm is perhaps the most popular clustering method for scRNA-seq, and was popularized by Seurat.

It can be parametrized with different resolutions that determine how large communities are detected in the SNN graph. Generally, lower resolutions result in coarse-grained clusters, while higher ones in more fine-grained structures.

The relationships in cluster abundances under different resolutions are visualized in the clustree plot below. Stable clusters across different resolutions can be quickly find as straight or little branched vertical lines.

r scdrake::create_a_link(cluster_graph_louvain_clustree_file, "**PDF with clustree**", href_rel_start = fs::path_dir(report_html_file))

r scdrake::format_used_functions(c("igraph::cluster_louvain()", "clustree::clustree()"))

res <- scdrake::generate_dimred_plots_clustering_section(
  dimred_plots_clustering_files, dimred_plots_clustering_united_files, "graph", "louvain", fs::path_dir(report_html_file), 4
)

bioinfocz/scdrake documentation built on Jan. 29, 2024, 10:24 a.m.