NAs in "target" and "ligand_target_weighted" columns `ligand_activities_targets_DEgenes$ligand_activities` dataframe

[browaeysrobin]

NAs in "target" and "ligand_target_weighted" columns `ligand_activities_targets_DEgenes$ligand_activities` dataframe

[browaeysrobin]

Hi @SergioRodLla

These NA values indicate that for a given ligand, no target gene was found. Meaning that none of the genes in the geneset of interest (=DE genes) is in the top_n (default 250) potential target genes of that ligand.
Using the adjusted p-values as cutoff to define the geneset of interest / DE genes will sometimes lead to a very small nr of genes. As a consequence, the chances of not finding target genes back increases, leading to more rows with NA values.

For inferring target genes and predicting ligand activities it is always recommended to have a sufficient nr of genes in the genesets of interest (some guidelines: > 20 & <2000).