Abstract
The available sequence data of RNA molecules have greatly increased in the past years. Unfortunately, while computational power is still under exponential growth, the computer prediction quality from sequence to final structure is still inferior to labour-intensive experimental work. Although a reliable end-to-end procedure has already been developed for proteins since Alphafold2, while its successor AlphaFold3 can also predict RNA, its confidence, in particular for novel sequences and folds, still appears limited. Another strategy entails two steps: (i) predicting potential contacts in the form of a contact map from evolutionary data; and (ii) simulating the molecule with a physical force field while using the contact map as restraint. However, the quality of the structure prediction crucially depends on the quality of the contact map. Until now, only the proportion of true positive contacts was considered as a quality characteristic. We propose to also include the distribution of these contacts, and have done so in our recent studies. We observed that the clustering of contacts, as is common for many artificial intelligence algorithms, has a negative impact on prediction quality. In contrast, a more distributed topology is beneficial. We have applied these findings from computer experiments to current algorithms and introduced a measure of distribution, the Gaussian score.