In a recent study posted to the bioRxiv* preprint server, researchers used machine learning (ML) tools to find animal coronaviruses (CoVs), each alpha and beta CoVs, previously unknown to contaminate humans.

Study: Using machine learning to detect coronaviruses potentially infectious to humans. Image Credit: MAVV/Shutterstock

Background

It has remained difficult to predict which animal CoVs might infect humans because their whole host range is unknown. As an illustration, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) originated in an animal host, most definitely bats. After a number expansion event, an important step in viral evolution, SARS-CoV-2 spilled over into humans. Thus, it’s crucial to survey all alpha and beta CoVs that infect animals near humans (e.g., cattle, comparable to pigs) that facilitate their zoonotic transmission.

Each alignment-based and alignment-free approaches have shown promise when addressing the problem of viral host prediction, however the former exhibits poor efficiency because the sequence lengths increase. Likewise, alignment-free methods don’t account for the relative position of the amino acid (AA) residues across the sequence.

In regards to the study

In the current study, researchers developed a novel machine-learning model to predict the binding between the spike (S) protein of alpha and beta CoVs and a human receptor, comparable to human dipeptidyl-peptidase 4 (hDPP4) and angiotensin-converting enzyme 2 (ACE2).

To this end, they first downloaded 28,368 spike (S) protein sequences of all alpha and beta CoVs from the National Center for Biotechnology Information Virus database. They used a skip-gram model to convert this data into vectors that encoded the association between adjoining length k protein sequences called k-mers. Next, a classifier used these vectors to attain each protein sequence per its human receptor binding potential, known as the human-Binding Potential (h-BiP).

The ultimate alpha and beta CoV dataset spanning all their clades and variants had 2,534 AA sequences, based on which there have been 1705 and 829 viruses with positive and negative annotations for human binding, respectively. Thus, the researchers split these 2,534 AA sequences right into a training (85%) and test set (15%).

Further, the researchers used a subset of 424 sequences to generate a phylogenetic tree for the S protein of alpha and beta CoVs. The team used starting receptor-binding domain (RBD) structures of LYRa3 and LYRa11, generated using AlphaFold, for molecular dynamics (MD) simulations. The MD package YASARA helped simulate protein-protein interactions by substituting individual AA residues and trying to find minimum-energy conformations on the ultimate modified candidate structures. The team also performed an energy minimization (EM) routine for all modified candidate structures until free energy stabilized to inside 50 Joules/mol. Resulting from the high accuracy of the classifier, the h-BiP rating correlated with the percent sequence identity (in %) against human viruses. The team computed the pairwise % sequence identity between all seven human CoVs and the S protein sequences within the study dataset to pick out the utmost for every. Notably, all viruses with ≥97 % identity with previously known human CoVs had an h-BiP rating >0.5.

Notably, the h-BiP rating detected binding in cases of low sequence identity and discriminated between the binding potential for viruses with nearly the identical sequence identity.

Results and conclusion

The researchers discovered LYRa326 and Bt13325, two viruses whose human binding properties are yet unknown, though they’d high h-BiP scores. In support, phylogenetic evaluation revealed that these two viruses were related to non-human CoVs previously known to bind to human receptors. The receptor binding motifs (RBM) inside the receptor binding domain (RBD) of the S protein is available in direct contact with the host receptor. The multiple sequence alignment of the RBMs of Bt133 and LYRa3 with related viruses uncovered that they conserve contact residues that interact with the human receptor(s).

As an illustration, Bt133 had conserved all its eight contact residues utilized by Tylonycteris bat CoV HKU4 (Ty-HKU4) to bind hDPP4  despite having 13 RBD mutations. Similarly, LYRa3, phylogenetically related to SARS-CoV Tor2, had conserved 12 of its 17 contact residues that bind to hACE2. Furthermore, aside from residue 441, it had equivalent sequences on the RBD. MD simulations of the RBD further validated this binding and identified contact residues that sure human receptors.

Finally, the researchers tested whether this model surveyed host expansion events. They emulated the conditions before SARS-CoV-2 advent by removing all SARS-CoV-2 S protein sequences from the training set. They found that the re-trained ML model successfully predicted the binding between a human receptor and the wild-type SARS-CoV-2 S, with an h-BiP rating equal to 0.96. Overall, the proposed ML-based method could prove to be a helpful tool for detecting, from an unlimited pool of animal CoVs, which viruses could cross species-barrier to contaminate humans.

*Essential notice

bioRxiv publishes preliminary scientific reports that aren’t peer-reviewed and, subsequently, shouldn’t be considered conclusive, guide clinical practice/health-related behavior, or treated as established information.