For impatient people
The tool must be executed on a Conda envirnoment which can be installed following the steps reported in the PREREQUISITES section of this manual.
SCALT requires a scRNA read counts matrix, having the genes on the rows and the cells ids on the columns, and the gene notation reported in the table, either gene_symbol or ensembl_id. The program for classification is called SCALT.py
SCALT classifies each cell to one of the 471 cell types available using the followinmg command:
python3 SCALT.py read_counts.py -Notation ensembl_id
The outpus are:
a REPORT.html file reporting a collection of plots that summarize the results of the experiment;
a directory called results_directory which contains all the metadata originated from the analysis.