standardise_ctd will use orthogene to convert non-human genes to 1:1 human orthologs (can be any species, not just muse).
If CTD is from human
standardise_ctd has the ability to detect ensembl IDs and convert them to HGNC symbols using orthogene. Since there's often more Ensembl IDs than HGNC symbols, orthogene will aggregate aggregate the matrix by summing read counts.
However, when the input to the generate_celltype_data has (non)-human Ensembl IDs, it would be nice to aggregate these within species first (as above), then proceed to dropping non 1:1 orthologs.
standardise_ctdstandardise_ctdis now called withinbootstrap_enrichment_test. https://github.com/bschilder/EWCE/blob/8e9ca154c98f2afb4c01a0a0e1ade92a690a2cce/R/bootstrap_enrichment_test.R#L134If CTD is from non-human
standardise_ctdwill useorthogeneto convert non-human genes to 1:1 human orthologs (can be any species, not just muse).If CTD is from human
standardise_ctdhas the ability to detect ensembl IDs and convert them to HGNC symbols usingorthogene. Since there's often more Ensembl IDs than HGNC symbols,orthogenewill aggregate aggregate the matrix by summing read counts.https://github.com/bschilder/EWCE/blob/8e9ca154c98f2afb4c01a0a0e1ade92a690a2cce/R/extract_matrix.R#L74
generate_celltype_dataIn addition, I've embedded
standardise_ctddirectly withingenerate_celltype_data. It can now convert to 1:1 human HGNC symbols. https://github.com/bschilder/EWCE/blob/8e9ca154c98f2afb4c01a0a0e1ade92a690a2cce/R/generate_celltype_data.r#L110However, when the input to the
generate_celltype_datahas (non)-human Ensembl IDs, it would be nice to aggregate these within species first (as above), then proceed to dropping non 1:1 orthologs.