Reliability with Single Source and Sink Samples

[jianglory]

Dear FEAST group,

I am currently using FEAST for microbial source tracking and have a couple of questions:

Reliability with Single Source and Sink Samples: Is the output from Feast reliable when there is only one sample in the source and one in the sink? I am particularly interested in understanding the tool’s performance under this minimal sample scenario.

Variance in Source Contribution with Additional Samples: I observed a significant change in the proportion attributed to a source sample (say, Sample A) in the sink when I added an extra source sample (for instance, Sample B) compared to when only one source sample (Sample A) was provided. Could you provide some guidance on how to interpret these results or suggest best practices for situations where the number of source samples varies?

Any insights or recommendations you can provide would be greatly appreciated.

Thank you very much!