DESeq estimateDispersions warning messages

[atkinson]

I'm using DESeq to compare one condition with 2 biological replicates, with another condition which has only one replicate.

When running estimateDispersions I get the following warning messages:

Warning messages:
1: In log(ifelse(y == 0, 1, y/mu)) : NaNs produced
2: step size truncated due to divergence

I'm not too sure what these mean, and do I need to worry about them?

Would be very grateful if anyone has encountered these before and is able to tell me what they mean.

Thanks,

Sophie

[carmeyeii]

Hi Sophie,

Did you manually enter the sizeFactors?

If so, I think this is an error that can occur when some sizeFactors are >1.