An official website of the United States government
Protein degradation through the Ubiquitin (Ub)-26S Proteasome System (UPS) is a major gene expression regulatory pathway in plants. In this pathway, the 76-amino acid Ub proteins are covalently linked onto a large array of UPS substrates with the help of three enzymes (E1 activating, E2 conjugating, and E3 ligating enzymes) and direct them for turnover in the 26S proteasome complex. The S-phase Kinase-associated Protein 1 (Skp1), CUL1, F-box (FBX) protein (SCF) complexes have been identified as the largest E3 ligase group in plants due to the dramatic number expansion of the FBX genes in plant genomes. Since it is the FBX proteins that recognize and determine the specificity of SCF substrates, much effort has been done to characterize their genomic, physiological, and biochemical roles in the past two decades of functional genomic studies. However, the sheer size and high sequence diversity of the FBX gene family demands new approaches to uncover unknown functions. In this work, we first identified 82 known FBX members that have been functionally characterized up to date in Arabidopsis thaliana . Through comparing the genomic structure, evolutionary selection, expression patterns, domain compositions, and functional activities between known and unknown FBX gene members, we developed a neural network machine learning approach to predict whether an unknown FBX member is likely functionally active in Arabidopsis, thereby facilitating its future functional characterization.
more »
« less
Genomic language model predicts protein co-regulation and function
<bold>Abstract</bold> 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.
more »
« less
- Award ID(s):
- 2032259
- PAR ID:
- 10498809
- Publisher / Repository:
- Nature Publishing Group
- Date Published:
- Journal Name:
- Nature Communications
- Volume:
- 15
- Issue:
- 1
- ISSN:
- 2041-1723
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Chloroviruses (family Phycodnaviridae) are dsDNA viruses found throughout the world’s inland waters. The open reading frames in the genomes of 41 sequenced chloroviruses (330 ± 40 kbp each) representing three virus types were analyzed for evidence of evolutionarily conserved local genomic “contexts”, the organization of biological information into units of a scale larger than a gene. Despite a general loss of synteny between virus types, we informatically detected a highly conserved genomic context defined by groups of three or more genes that we have termed “gene gangs”. Unlike previously described local genomic contexts, the definition of gene gangs requires only that member genes be consistently co-localized and are not constrained by strand, regulatory sites, or intervening sequences (and therefore represent a new type of conserved structural genomic element). An analysis of functional annotations and transcriptomic data suggests that some of the gene gangs may organize genes involved in specific biochemical processes, but that this organization does not involve their coordinated expression.more » « less
-
Using Network-Based Machine Learning to Predict Transcription Factors Involved in Drought Resistancenull (Ed.)Gene regulatory networks underpin stress response pathways in plants. However, parsing these networks to prioritize key genes underlying a particular trait is challenging. Here, we have built the Gene Regulation and Association Network (GRAiN) of rice ( Oryza sativa ). GRAiN is an interactive query-based web-platform that allows users to study functional relationships between transcription factors (TFs) and genetic modules underlying abiotic-stress responses. We built GRAiN by applying a combination of different network inference algorithms to publicly available gene expression data. We propose a supervised machine learning framework that complements GRAiN in prioritizing genes that regulate stress signal transduction and modulate gene expression under drought conditions. Our framework converts intricate network connectivity patterns of 2160 TFs into a single drought score. We observed that TFs with the highest drought scores define the functional, structural, and evolutionary characteristics of drought resistance in rice. Our approach accurately predicted the function of OsbHLH148 TF, which we validated using in vitro protein-DNA binding assays and mRNA sequencing loss-of-function mutants grown under control and drought stress conditions. Our network and the complementary machine learning strategy lends itself to predicting key regulatory genes underlying other agricultural traits and will assist in the genetic engineering of desirable rice varieties.more » « less
-
Abstract Gene duplication is a fundamental part of evolutionary innovation. While single-gene duplications frequently exhibit asymmetric evolutionary rates between paralogs, the extent to which this applies to multi-gene duplications remains unclear. In this study, we investigate the role of genetic context in shaping evolutionary divergence within multi-gene duplications, leveraging microsynteny to differentiate source and target copies. Using a dataset of 193 mammalian genome assemblies and a bird outgroup, we systematically analyze patterns of sequence divergence between duplicated genes and reference orthologs. We find that target copies, those relocated to new genomic environments, exhibit elevated evolutionary rates compared to source copies in the ancestral location. This asymmetry is influenced by the distance between copies and the size of the target copy. We also demonstrate that the polarization of rate asymmetry in paralogs, the “choice” of the slowly evolving copy, is biased towards collective, block-wise polarization in multi-gene duplications. Our findings highlight the importance of genetic context in modulating post-duplication divergence, where differences in cis-regulatory elements and co-expressed gene clusters between source and target copies may be responsible. This study presents a large-scale test of asymmetric evolution in multi-gene duplications, offering new insight into how genome architecture shapes functional diversification of paralogs. Significance statementAfter a gene is duplicated, reduced selective constraints can lead the two copies to rapidly diverge, with one copy often evolving faster and occasionally gaining a new function. We quantify the influence of genetic context in choosing which copy of a duplicated gene has an elevated substitution rate. In a representative dataset of 193 mammalian genomes, we found strong evidence that gene copies pasted into new genomic locations tend to evolve faster than the corresponding copies in ancestral locations, suggesting an important role for the regulatory environment. The asymmetry in evolutionary rates of duplicated genes persists even for very large multigenic duplications, up to the scale of megabases, indicating that regulatory interactions frequently reach farther than previously thought.more » « less
-
Martelli, Pier Luigi (Ed.)Abstract Motivation Clustering spatial-resolved gene expression is an essential analysis to reveal gene activities in the underlying morphological context by their functional roles. However, conventional clustering analysis does not consider gene expression co-localizations in tissue for detecting spatial expression patterns or functional relationships among the genes for biological interpretation in the spatial context. In this article, we present a convolutional neural network (CNN) regularized by the graph of protein–protein interaction (PPI) network to cluster spatially resolved gene expression. This method improves the coherence of spatial patterns and provides biological interpretation of the gene clusters in the spatial context by exploiting the spatial localization by convolution and gene functional relationships by graph-Laplacian regularization. Results In this study, we tested clustering the spatially variable genes or all expressed genes in the transcriptome in 22 Visium spatial transcriptomics datasets of different tissue sections publicly available from 10× Genomics and spatialLIBD. The results demonstrate that the PPI-regularized CNN constantly detects gene clusters with coherent spatial patterns and significantly enriched by gene functions with the state-of-the-art performance. Additional case studies on mouse kidney tissue and human breast cancer tissue suggest that the PPI-regularized CNN also detects spatially co-expressed genes to define the corresponding morphological context in the tissue with valuable insights. Availability and implementation Source code is available at https://github.com/kuanglab/CNN-PReg. Supplementary information Supplementary data are available at Bioinformatics online.more » « less
