Deep Learning for Genomics: From Early Neural Nets to Modern Large Language Models

Abstract

The data explosion driven by advancements in genomic research, such as high-throughput sequencing techniques, is constantly challenging conventional methods used in genomics. In parallel with the urgent demand for robust algorithms, deep learning has succeeded in various fields such as vision, speech, and text processing. Yet genomics entails unique challenges to deep learning, since we expect a superhuman intelligence that explores beyond our knowledge to interpret the genome from deep learning. A powerful deep learning model should rely on the insightful utilization of task-specific knowledge. In this paper, we briefly discuss the strengths of different deep learning models from a genomic perspective so as to fit each particular task with proper deep learning-based architecture, and we remark on practical considerations of developing deep learning architectures for genomics. We also provide a concise review of deep learning applications in various aspects of genomic research and point out current challenges and potential research directions for future genomics applications. We believe the collaborative use of ever-growing diverse data and the fast iteration of deep learning models will continue to contribute to the future of genomics.

Large language models reshaping molecular biology and drug development

“…Genomic data is packed with information that has the potential to reveal previously unknown aspects of human health and disease (Bustamante et al, 2011). LLMs are able to examine and understand genomic sequences, which enables them to find genetic differences and determine the possible influence these variations may have on disease susceptibility, medication response, and treatment results (Consens et al, 2023;Yue et al, 2023). These models can potentially assist in identifying disease-causing mutations, unraveling complicated genetic relationships, and guiding precision medicine techniques that are customized to the specific needs of patients (Benegas et al, 2023).…”

SegmentNT: annotating the genome at single-nucleotide resolution with DNA foundation models

“…The intersection of genomics research and deep learning methods is profoundly changing our ability to understand the information encoded in each of the 3 billion nucleotides in the human genome and to accurately assess their influence with respect to different gene-regulatory activity layers, ranging from regulatory elements and transcriptional activation to splicing and polyadenylation [1,2]. Sequence-based machine learning models trained on large-scale genomics data capture complex patterns in the sequence and can predict diverse molecular phenotypes with great accuracy.…”