I have always had a doubt

[tzjsdss]

I have always had a doubt, or maybe I didn't read the paper very clearly
When searching on the web version, a 0-1 Probe will appear Value, how does this number reflect the results, and generally speaking, what are the scores above that are reliable? 0.6？ 0.3？

[martin-steinegger]

The probability score is only relevant for hits with an E-value greater than 0.1. It is a fitted score calibrated to a FP probability threshold. For more information, we have provided a brief guide on the probability score here: Foldseek Wiki - What is the hit probability?. I hope this helps!

[tzjsdss]

Thank you very much for your answer. I'm sorry that your answer has left me even more confused. I have browsed the wiki, but I still cannot understand the impact of p-value size on the results. If possible, could you please answer the following questions

1. Can I just look at the Prob directly To determine the similarity between two proteins based on their results
2. If possible, what is the reference range<0.3? 0.6>?
   If not possible, I should also pay close attention to the comparison results Seq Id？. E-Value？ What is the score in comparison to p-value in the wiki? Is it TM score?