1-million count requirement too strict?

[ashipde]

I feel that the 1-million requirement for total miRNA counts is too strict. I have some datasets where none or only a few of the samples meet this criteria. Perhaps the minimum count value can be made a user setting.

[mxhp75]

Hi Ashipde

Thank you for raising this important issue about the 1-million read threshold in our miRNA haemolysis detection app.

I understand the practical challenge you're facing with datasets where few or no samples meet this threshold. The 1-million read requirement was established during our development and validation process, where we found that the geometric mean calculations and the differential expression between haemolysis-associated miRNAs and background miRNAs became less reliable below this threshold.

That said, I appreciate that real-world samples don't always yield ideal read counts, and researchers still need tools to assess potential haemolysis. Here's what I propose:

1. In the next update, I will implement a user-adjustable threshold with appropriate warnings about reduced reliability at lower read depths.

2. For transparency, I will add documentation explaining that results from samples with <1M reads should be interpreted with caution, as they fall outside our validated parameters.

Thanks again for your feedback - it's input like this that helps improve scientific tools for the entire community.