Year: 2024
Type: article
Abstract: Deciphering the relationship between a gene and its genomic context is fundamental to understanding and engineering biological systems. Machine learning has shown promise in learning latent relationships underlying the sequence-structure-function paradigm from massive protein sequence datasets. However, to date, limited attempts have been made in extending this continuum to include higher order genomic context information. Evolutionary processes dictate the specificity of genomic contexts in which a gene is found across phylogenetic distances, and these emergent genomic patterns can be leveraged to uncover functional relationships between gene products. Here, we train a genomic language model (gLM) on millions of metagenomic scaffolds to learn the latent functional and regulatory relationships between genes. gLM learns contextualized protein embeddings that capture the genomic context as well as the protein sequence itself, and encode biologically meaningful and functionally relevant information (e.g. enzymatic function, taxonomy). Our analysis of the attention patterns demonstrates that gLM is learning co-regulated functional modules (i.e. operons). Our findings illustrate that gLM's unsupervised deep learning of the metagenomic corpus is an effective and promising approach to encode functional semantics and regulatory syntax of genes in their genomic contexts and uncover complex relationships between genes in a genomic region.
Source: Nature Communications
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: gold
Estimated APC paid: $6,790
Awards: Collaborative Research: A multidimensional approach to understanding microbial carbon cycling beneath the seafloor during cool hydrothermal circulation, Towards Life with a Reduced Protein Alphabet, G4768352037, G5304168789, Exploring the unknown protein universe using evolutionary information, HOW DO WE CONTEND WITH THE TRULY ALIEN? MOST STRATEGIES FOR BIOSIGNATURE DETECTION RELY UPON FINDING FEATURES KNOWN TO BE ASSOCIATED WITH TERRESTRIAL LIFE SUCH AS PARTICULAR CLASSES OF MOLECULES BUT LIFE MAY BE VASTLY DIFFERENT ON OTHER WORLDS. THIS WORK WILL FURTHER DEVELOP SEVERAL TRULY AGNOSTIC APPROACHES NONE OF WHICH PRESUPPOSE A PARTICULAR MOLECULAR FRAMEWORK. SOME OF THESE AREAS WILL INVOLVE NEW LINES OF INQUIRY THAT CAN BE ACCOMPLISHED USING INSTRUMENTATION THAT IS EITHER HIGH HERITAGE OR ON A PATH TOWARD HIGH TRL. OTHERS WILL NECESSITATE THE DESIGN OF COMPLETELY NEW ANALYTICAL TECHNIQUES. OUR NAI WILL NOT ONLY ADVANCE EACH OF THESE AGNOSTIC DETECTION CONCEPTS SEPARATELY BUT ALSO JOIN THESE TRADITIONALLY FRAGMENTED EFFORTS IN A UNIFIED RESEARCH PROGRAM ONE THAT DRAWS LESSONS FROM OUR OPERATIONAL EXPERIENCES IN PLANETARY EXPLORATION THUS FAR. WE WILL BRING TOGETHER THEORETICIANS AND EXPERIMENTALISTS AS AN INCUBATOR FOR NEW AGNOSTIC LIFE DETECTION APPROACHES CREATE A COMMON TRAINING SET OF BIOTIC AND ABIOTIC TEST SUBSTRATES ESTABLISH LIKELIHOODS AND THRESHOLDS WITH PROBABILISTIC MODELS AND DEVELOP A BROADER FORMULATION OF THE LADDER OF LIFE. IN THE PROCESS WE WILL MERGE THE BROAD AND NARROW ENDS OF THE LIFE DETECTION SPECTRUM INTO A COLLECTIVE DIALOGUE ONE THAT WILL HELP US SYNTHESIZE PRIORITIZE EVALUATE AND STRENGTHEN THE SEARCH FOR LIFE ON FUTURE SPACE MISSIONS.
OpenAlex ID: W4393870943