Figure 4: CASP15 Protein assembly subunit interface agreement with experiment. Bars show the best agreement obtained between computed and experimental interfaces, using the Interface Contact Score (ICS). Green bars indicate interfaces in targets where there are homologous structures for all component subunits and interfaces. Blue bars are those where there are homologous structures for some subunits and interfaces, and red those where there is no structural homology. More than half the interfaces reach a score of 0.8 or higher, and so may approach experimental accuracy. About a quarter reach a more stringent criterion of ICS > 90.
In CASP15 a total of 40 protein assembly targets were assessed, ranging from simple obligate dimers to very large complexes with up to 11 polypeptide chains and up to more than 4000 amino acids. Figure 4 shows the distribution of interface structure agreement with experiment in terms of the Interface Contact score (ICS), the F1 measure of contact agreement introduced earlier. Examination of interfaces with lowest ICS values ( <0.6) is informative. Three of the eight targets bearing the poorly modeled interfaces are immune complexes, discussed below. Four interfaces are from large complexes, which tend to be a slightly lower accuracy overall, and one is for a viral coat protein homodimer (T1123). The multiple sequence alignment for the monomer of this homo-dimeric target is shallow (10 sequences), but the subunit’s best models are of high accuracy (best GDT_TS of 87), so it is unclear why this interface has poor agreement with experiment.