Interface predicted template modelling (ipTM)

Measure of the accuracy of the predicted structure of a multimer, used by AlphaFold and Boltz. It measures the accuracy of the predicted interface between the subunits of the protein-protein complex.

ipTM measures the accuracy of the predicted relative positions of the subunits forming the protein-protein complex. Values higher than 0.8 represent confident high-quality predictions, while values below 0.6 suggest likely a failed prediction. ipTM values between 0.6 and 0.8 are a grey zone where predictions could be correct or wrong. These values assume modelling with multiple recycling steps, so the process of prediction reaches a degree of convergence. In large-scale screenings for protein-protein interactions, often settings optimised for the speed of prediction are used, e.g. very few or no recycling steps. In such cases ipTM thresholds as low as 0.3 have been used for initial screening; importantly though, all pairs of proteins with ipTM scores higher than 0.3 were then subjected to additional examination (e.g. Weeratunga et al., 2023). Disordered regions and regions with low pLDDT score may negatively impact the ipTM score even if the structure of the complex is predicted correctly.

ipTM may be more useful to users than pTM. This is because the quality of the prediction of the relative positions of the subunits and the quality of the whole complex prediction are highly interdependent: if the relative positions of the subunits are correct (as reflected in a high ipTM score), users can expect that the whole complex is also correct.

This card will indicate the value when using Aura and the Fold workflow.