BLAST data output, mismatch and gap values

Question

[hollie]

I have generated some genomic sequencing data and ran the files through BLAST to classify the genome. I now want to filter the data output using the mismatch and gap values but I am not sure what values to use for each. I want to eliminate any reads that have been classified with a lot of gaps or mismatches as these are likely to not be accurate. Does anybody have an idea of what values to use to subset the data based on mismatches and gaps as I cannot find any suggestions in the literature. Thanks!