I have an RNA experiement with 2 factors, and one is confounding with the batch effect. Running ComBat seq on my data, with a formula of ComBat_seq(count_matrix, batch=batch, group = group), with the factor that is not a linear combination of the batch effect, results in a poor separation of my batches(only 2 out of 6 batches are separated). Running ComBat_seq(count_matrix, batch=batch) does still not yield a good separation on the data.
I have managed to obtained what I think it's a good separation of my data, only after running two consecutive corrections on the raw counts of the type: ComBat_seq(count_matrix, batch=batch),followed by ComBat_seq(count_matrix, batch=batch, group = group). The same batch is used for the second correction.
My question is, is this double adjusted data usable for any DE genes analysis?
I have an RNA experiement with 2 factors, and one is confounding with the batch effect. Running ComBat seq on my data, with a formula of ComBat_seq(count_matrix, batch=batch, group = group), with the factor that is not a linear combination of the batch effect, results in a poor separation of my batches(only 2 out of 6 batches are separated). Running ComBat_seq(count_matrix, batch=batch) does still not yield a good separation on the data. I have managed to obtained what I think it's a good separation of my data, only after running two consecutive corrections on the raw counts of the type: ComBat_seq(count_matrix, batch=batch),followed by ComBat_seq(count_matrix, batch=batch, group = group). The same batch is used for the second correction.
My question is, is this double adjusted data usable for any DE genes analysis?