search

BIMIB-DISCo / scFBA

6 stars 5 forks source link

Running scFBA without bulk/pooled sample data #5

Open riddhimandhar opened 3 years ago

riddhimandhar commented 3 years ago

Hi, Is it possible to run scFBA without including pooled/bulk RNA-seq data? (Because sometimes this data is not available). From your README file, it seems that it is possible to run scFBA without bulk data. Is this correct? If yes, how should we format the dataset file? Should we replace the pooled values by NA/NaN?

Thank you very much for your help.

Regards, RDhar

DavideMaspero commented 3 years ago

Dear Dr. RDhar,

sorry for the very late reply.

scFBA is conceived to use the information from a pooled/bulk sequencing experiment to compute statistics on the single cell gene expression levels. There is not a explicit way to not provide the pooled/bulk profile. Anyway, you could try by passing the averege expression level of the single cell sequencing experiment. For example, the following command can replace the line 25 of the scFBA_test script BC04 = makeSCdataset(mean(BC04_filt{:,7:end}', 'omitnan'), BC04_filt{:,7:end}, BC04_filt.Properties.VariableNames(7:end), BC04_filt.HGNC_ID, 10^-4);

Best regards, Davide Maspero