I recently searched the same plasma file with DIANN1.9 and DIANN1.9.1 and found that 1.9 identified 3800 proteins and 1.9.1 identified 7999 proteins. May I ask why this is,Attached are logs that run in both versions,I look forward to hearing from you, thank you very much
Need to address the warning, i.e. generate an in silico-predicted library in a separate pipeline step and then use it to process the raw data, now without activating FASTA digest.
I recently searched the same plasma file with DIANN1.9 and DIANN1.9.1 and found that 1.9 identified 3800 proteins and 1.9.1 identified 7999 proteins. May I ask why this is,Attached are logs that run in both versions,I look forward to hearing from you, thank you very much
1.9 log.docx 1.9.1 log.docx