Low read depths in MAF file record should be excluded from analysis?

Question

[Serge]

Hello, I am a bioinformatics newbie analysing MAF file format data with t_depth and n_depth fields. My understanding is that low read count values for these fields are undesirable (false positives, lack of accuracy,etc). Is there a standard technique or process to establish a cutoff value to exclude data with read counts below that cutoff value?

I use R for this work, so an R package solution would be ideal, but happy to do it another way if required.

Update: I would appreciate being updated if this is such a stupid question that it's not worth answering. Hard to know if you are completely wrong sometimes.