Abstract
Genomes have an inherent context dictated by the order in which the nucleotides and higher order genomic elements are arranged in the DNA/RNA. Learning this context is a daunting task, governed by the combinatorial complexity of interactions possible between ordered elements of genomes. Can natural language processing be employed on these orderly, complex and also evolving datatypes (genomic sequences) to reveal the latent patterns or context of genomic elements (e.g Mutations)? Here we present an approach to understand the mutational landscape of Covid-19 by treating the temporally changing (continuously mutating) SARS-CoV-2 genomes as documents. We demonstrate how the analogous interpretation of evolving genomes to temporal literature corpora provides an opportunity to use dynamic topic modeling (DTM) and temporal Word2Vec models to delineate mutation signatures corresponding to different Variants-of-Concerns and tracking the semantic drift of Mutations-of-Concern (MoC). We identified and studied characteristic mutations affiliated to Covid-infection severity and tracked their relationship with MoCs. Our ground work on utility of such temporal NLP models in genomics could supplement ongoing efforts in not only understanding the Covid pandemic but also provide alternative strategies in studying dynamic phenomenon in biological sciences through data science (especially NLP, AI/ML).
Competing Interest Statement
Authors are salaried scientists at TCS Research, Tata Consultancy Services Ltd. SN is currently an industry sponsored PhD fellow at CSIR IGIB.
Funding Statement
Authors are salaried employees at TCS Research, Tata Consultancy Services Ltd India
Author Declarations
I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.
Yes
The details of the IRB/oversight body that provided approval or exemption for the research described are given below:
GISAID required sharing of a acknowledgement document pertaining to data contributors on GISAID platform on which this research is based. The acknowledgement document is provided as Supplementary File 2 in compliance with the same.
I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.
Yes
I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).
Yes
I have followed all appropriate research reporting guidelines and uploaded the relevant EQUATOR Network research reporting checklist(s) and other pertinent material as supplementary files, if applicable.
Yes
Data Availability
All data produced in the present work are contained in the manuscript. Genome sequences and metadata employed in this research are available on gisaid.org