Bioinformatic assessment of the potential amyloidogenicity of the human and evolutionarily more ancient proteomes.

AmyloGram is a computer program that uses n-gram encoding and a random forest classifier to produce a numerical score between 0 and 1 for the predicted amyloidogenicity of a given protein sequence. In a variety of recent studies, we have used AmyloGram to obtain an overall amyloidogenicity score for members of the human proteome. Of 83,567 full-length canonical human polypeptides, 79.2% had a score exceeding 0.7 (the median was 0.813), consistent with the view that most natural protein sequences contain elements that are in fact potentially amyloidogenic. Here, we first asked whether this operational threshold is supported by orthogonal predictors and curated amyloid proteins, and then whether similarly high scores are also observed in evolutionarily ancient proteomes. For the human proteome, PASTA2 values correlated positively with AmyloGram scores (r2 = 0.374 for minimum free energy and 0.321 for average free energy), and proteins with AmyloGram scores ≥0.7 were significantly enriched for strongly negative PASTA2 values (for average free energy <-10 PEU: 4724/66190 versus 75/17,377; χ2P = 2.4 × 10-250). In AmyPro, 117 curated amyloid proteins had substantially higher median AmyloGram scores than did the ~8 curated non-amyloids, although the imbalance of this dataset demands caution. AMYPred-FRL showed only a weak and partly discordant relationship with AmyloGram in the archaeal test proteome examined. We then computed AmyloGram score distributions for the proteomes of 130 other organisms, including archaea, Gram-negative bacteria, Gram-positive bacteria and viruses, representing 475,999 proteins in total. The corresponding organism-level median AmyloGram scores by domain were 0.822, 0.853, 0.851, and 0.825, respectively, while the overall median across all 130 organism-level medians was 0.828. By contrast, 1000 random 100-mer sequences generated with equal amino-acid probabilities had a median AmyloGram score of 0.740. Weighting according to known residue distributions did not change this. However, a weak dependence of AmyloGram score on sequence length within an organism was observed, and the median length of human proteins (∼375 residues) gave a median AmyloGram score of 0.837. Taken together, these findings are consistent with, though they cannot prove, an early evolutionary origin of widespread amyloidogenic potential in natural proteins, and most likely reflect the biophysical properties of the amino acid residues that tend to induce amyloidogenesis.