Closed jefferyUstc closed 3 years ago
From the paper: "Counts file. scRNA-seq count data containing gene expression values in which rows are genes presented with gene names identifiers (Ensembl IDs, gene names or hgnc_symbol annotation) and columns are cells. We recommend using normalized count data. Importantly, the user needs to specify whether the data was log-transformed when using the subsampling option. "
Hi, cellphonedb team, I'd like to know what kind of data should feed as the input data. the raw counts, the normalized or log-normalized data?