Is there a minimum number of cells with a particular variant/edit to call something a true positive vs. a false positive?

Most false positive variants are present in <= 3 cells in a sample, but not all variants present in <= 3 cells are false positives. We’ve observed True positive variants in as low as 2 cells in the development testing for LOD. From past experiments, we’ve observed that most false positive variants are indels of maximum length of 2 bp present in less than 0.05% of cells as a Heterozygous variant. We recommend running a mock (negative control sample) to better understand false positive variants and to make a judgment call on variants seen in less than 0.1% of cells based on analysis of the negative control sample.