iiinova — where your digital self keeps evolving

Replies

ME/CFS Science @mecfsscience.org · 09/09/2025

1) There were some concerns that DecodeME might have suffered from population differences and that selection bias or different ancestry between patients and controls affected the DNA results. Looked closer into this and don't think this distorted the results. A brief thread 🧵

1124

ME/CFS Science @mecfsscience.org · 09/09/2025

2) The general principle to test for this is straightforward: with population differences, you expect p-value stats to be inflated across the board. If the signal reflects susceptibility to a disease, however, we expect a smaller proportion of p-values to be abnormal.

150