Skip to content

Conversation

@keiran-rowell-unsw
Copy link

@keiran-rowell-unsw keiran-rowell-unsw commented Jan 8, 2026

Test data for a MultiQC module for nf-core/proteinfold (draft PR here)

Covers the main metrics in simple .tsv format. See the EBI guide for the description of confidence scores common to all AI protein structure inference codes:

  • MSA - Multiple Sequence Alignment (for depth of homologues sequences)
  • pLDDT - predicted Local Distance Difference Test (spatial error per amino acid residue position)
  • pTM - predicted Template Modelling (global structure accuracy)
  • ipTM - interface score (predicting relative positional accuracy of the subunits)
  • PAE - [excluded from lightweight MultiQC report] predicted aligned error (matrix of relative positional error between every pair of residues)

@keiran-rowell-unsw keiran-rowell-unsw changed the title proteinfold v2 metrics from any protein structure prediction models New module: Proteinfold (protein structure AI inference programs - v2 metrics) Jan 8, 2026
@keiran-rowell-unsw keiran-rowell-unsw marked this pull request as draft January 11, 2026 22:49
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant