Assessing structure and disorder prediction tools for emerged proteins in the age of machine learning.

protein coding genes emerge from scratch in the non-coding regions of the genome and have, per definition, no homology to other genes. Therefore, their encoded proteins belong to the so-called "dark protein space". So far, only four protein structures have been experimentally approximated. Low homology, presumed high disorder and limited structures result in low confidence structural predictions for proteins in most cases. Here, we look at the most widely used structure and disorder predictors and assess their applicability for emerged proteins. Since AlphaFold2 is based on the generation of multiple sequence alignments and was trained on solved structures of largely conserved and globular proteins, its performance on proteins remains unknown. More recently, natural language models of proteins have been used for alignment-free structure predictions, potentially making them more suitable for proteins than AlphaFold2. We applied different disorder predictors (IUPred3 short/long, flDPnn) and structure predictors, AlphaFold2 on the one hand and language-based models (Omegafold, ESMfold, RGN2) on the other hand, to four de novo proteins with experimental evidence on structure. We compared the resulting predictions between the different predictors as well as to the existing experimental evidence. Results from IUPred, the most widely used disorder predictor, depend heavily on the choice of parameters and differ significantly from flDPnn which has been found to outperform most other predictors in a comparative assessment study recently. Similarly, different structure predictors yielded varying results and confidence scores for proteins. We suggest that, while in some cases protein language model based approaches might be more accurate than AlphaFold2, the structure prediction of emerged proteins remains a difficult task for any predictor, be it disorder or structure.