Hello,
Currently, I am running MEME with the multihit option. Since HyPhy Vision is not suitable for interpreting MEME-multihit results, I attempted to derive criteria for identifying “diversifying” selection by comparing the results from the web interface with the CSV files generated from the JSON outputs.
As my analyses using FEL, FUBAR, and SLAC on this dataset indicate purifying selection, I would like to pinpoint which codons exhibit episodic positive selection, and whether these signals are strong or negligible.
However, because the column headers in HyPhy Vision and the web interface results are not consistent, I am somewhat confused. I have checked the issue pages but have not been able to find a solution.
For example, in my dataset, codon 15 was classified as “diversifying.” My concern is that there is no p-value or p⁺ column reported for this site.
meanwhile, the hyphy vision column (which faithful to the .json headers), it mentioned the p+ and also the p value.
and I tried https://observablehq.com/@spond/meme to read the .json files
In my previous understanding, the p⁺ value can be used in combination with the p-value to determine whether β⁺ is significant. Therefore, the observablehq output seemed reasonable to me. Please let me know if this understanding is incorrect.
Therefore, is there any best practice for filtering out potential false positives or weak signals of positive selection from MEME results?
Thank you for the help!
Best,
Augustine
Hello,
Currently, I am running MEME with the multihit option. Since HyPhy Vision is not suitable for interpreting MEME-multihit results, I attempted to derive criteria for identifying “diversifying” selection by comparing the results from the web interface with the CSV files generated from the JSON outputs.
As my analyses using FEL, FUBAR, and SLAC on this dataset indicate purifying selection, I would like to pinpoint which codons exhibit episodic positive selection, and whether these signals are strong or negligible.
However, because the column headers in HyPhy Vision and the web interface results are not consistent, I am somewhat confused. I have checked the issue pages but have not been able to find a solution.
For example, in my dataset, codon 15 was classified as “diversifying.” My concern is that there is no p-value or p⁺ column reported for this site.
meanwhile, the hyphy vision column (which faithful to the .json headers), it mentioned the p+ and also the p value.
and I tried https://observablehq.com/@spond/meme to read the .json files
In my previous understanding, the p⁺ value can be used in combination with the p-value to determine whether β⁺ is significant. Therefore, the observablehq output seemed reasonable to me. Please let me know if this understanding is incorrect.
Therefore, is there any best practice for filtering out potential false positives or weak signals of positive selection from MEME results?
Thank you for the help!
Best,
Augustine