NetMHCpan: False peptide min/max boundaries

[jonasscheid]

The internal boundaries of netmhcpan regarding peptide lengths are 8 - 56. NetMHCpan 4.1 always picks the most relevant Core of the peptide and does prediction on it. That is currently prohibited by the supported length that is inherited by the netmhcpan 2.8 interface.

Regarding NetMHCIIpan there seems to be only a lower limit to 9mers, I can go beyond 100 amino acids, which clearly does not make sense. So we could also say that we restrict it to peptides having max length of 56

[christopher-mohr]

Thanks for adding the issue.

I'm not sure yet if I understand it completely. Could you provide an example how that results in different or wrong results? How does that affect the predictions when we provide peptides directly anyway?

I guess in general it makes sense though to adjust the allowed peptide lengths to the tools' values.

[jonasscheid]

I does not lead to wrong results, but class 1 peptides > 14 cannot be predicted by netmhcpan. Also netmhcIIpan is capped to peptides <20, which is a problem if we want to predict 21, 22mers. I think we don't need to change anything in parsing, it's just the boundaries of the peptide lengths.

What I wanted to say is that netmhc gets the most interesting core of the peptide and does the prediction on that:

[christopher-mohr]

I am still not sure I understand how the core is related to the problem with not having allowed all possible peptide lengths. The core is always the 9mer that is predicted to be in contact with the HLA molecule. I think the prediction is still done on the whole peptide though but would have to look up the details.

[jonasscheid]

You are right, sorry for the confusion. I checked prediction with the core peptide and the rank is different. So let's allow all possible peptide lengths then?

[christopher-mohr]

Sounds good to me, still worth to be changed. Thanks for bringing it up.