search

MannLabs / alphapeptdeep

Deep learning framework for proteomics
Apache License 2.0
101 stars 19 forks source link

Supported fragmentlosstype? #153

Open hxxhust163 opened 2 months ago

hxxhust163 commented 2 months ago

Hi

Thanks for your work in Alphapeptdeep. I found a tsv_str your provided in the /nbdev_nbs/pipeline_api.ipynb containing 'noloss, H3PO4, 1(+H2+O)1(+H3+O4+P)' in the FragmentLossType column. I wonder whether peptdeep supports the fragmentlloss such as 'H2O, NH3' which means loss of 18 and 17 in mass?

Thanks very much!

Best wishes Xiaoxiang

jalew188 commented 2 months ago

No, we did not train model for H2O and NH3 losses

jalew188 commented 2 months ago

In principle, it is very easy to train a new model for H2O and NH3 losses, I am just not sure if it is very necessary

hxxhust163 commented 2 months ago

OK, thanks again!

jalew188 commented 2 months ago

I will write a comprehensive tutorial about alphabase/alpharaw/peptdeep