scholarly journals DNABERT: pre-trained Bidirectional Encoder Representations from Transformers model for DNA-language in genome

2021 ◽  
Author(s):  
Yanrong Ji ◽  
Zhihan Zhou ◽  
Han Liu ◽  
Ramana V Davuluri
Keyword(s):  
Dna Sequences ◽  
Regulatory Elements ◽  
Ease Of Use ◽  
Fine Tuning ◽  
Supplementary Information ◽  
Sequence Motifs ◽  
Semantic Relationship ◽  
Accurate Identification ◽  
Conserved Sequence ◽  
Genome Wide

Abstract Motivation Deciphering the language of non-coding DNA is one of the fundamental problems in genome research. Gene regulatory code is highly complex due to the existence of polysemy and distant semantic relationship, which previous informatics methods often fail to capture especially in data-scarce scenarios. Results To address this challenge, we developed a novel pre-trained bidirectional encoder representation, named DNABERT, to capture global and transferrable understanding of genomic DNA sequences based on up and downstream nucleotide contexts. We compared DNABERT to the most widely used programs for genome-wide regulatory elements prediction and demonstrate its ease of use, accuracy and efficiency. We show that the single pre-trained transformers model can simultaneously achieve state-of-the-art performance on prediction of promoters, splice sites and transcription factor binding sites, after easy fine-tuning using small task-specific labeled data. Further, DNABERT enables direct visualization of nucleotide-level importance and semantic relationship within input sequences for better interpretability and accurate identification of conserved sequence motifs and functional genetic variant candidates. Finally, we demonstrate that pre-trained DNABERT with human genome can even be readily applied to other organisms with exceptional performance. We anticipate that the pre-trained DNABERT model can be fined tuned to many other sequence analyses tasks. Availability and implementation The source code, pretrained and finetuned model for DNABERT are available at GitHub (https://github.com/jerryji1993/DNABERT). Supplementary information Supplementary data are available at Bioinformatics online.

scholarly journals DNABERT: pre-trained Bidirectional Encoder Representations from Transformers model for DNA-language in genome

2020 ◽  
Author(s):  
Yanrong Ji ◽  
Zhihan Zhou ◽  
Han Liu ◽  
Ramana V Davuluri
Keyword(s):  
Dna Sequences ◽  
State Of The Art ◽  
Fine Tuning ◽  
Sequence Motifs ◽  
Semantic Relationship ◽  
Accurate Identification ◽  
Specific Data ◽  
Conserved Sequence ◽  
Conserved Sequence Motifs ◽  
Gene Regulatory

ABSTRACTDeciphering the language of non-coding DNA is one of the fundamental problems in genome research. Gene regulatory code is highly complex due to the existence of polysemy and distant semantic relationship, which previous informatics methods often fail to capture especially in data-scarce scenarios. To address this challenge, we developed a novel pre-trained bidirectional encoder representation, named DNABERT, that forms global and transferrable understanding of genomic DNA sequences based on up and downstream nucleotide contexts. We show that the single pre-trained transformers model can simultaneously achieve state-of-the-art performance on many sequence predictions tasks, after easy fine-tuning using small task-specific data. Further, DNABERT enables direct visualization of nucleotide-level importance and semantic relationship within input sequences for better interpretability and accurate identification of conserved sequence motifs and functional genetic variants. Finally, we demonstrate that pre-trained DNABERT with human genome can even be readily applied to other organisms with exceptional performance.

FindNonCoding: rapid and simple detection of non-coding RNAs in genomes

2021 ◽  
Author(s):  
Erik S Wright
Keyword(s):  
Pattern Mining ◽  
Ease Of Use ◽  
Supplementary Information ◽  
Sequence Motifs ◽  
Protein Coding ◽  
False Discovery ◽  
Non Coding Rna ◽  
Hairpin Loops ◽  
Simple Detection ◽  
Non Coding Rnas

Abstract Summary Non-coding RNAs are often neglected during genome annotation due to their difficulty of detection relative to protein coding genes. FindNonCoding takes a pattern mining approach to capture the essential sequence motifs and hairpin loops representing a non-coding RNA family and quickly identify matches in genomes. FindNonCoding was designed for ease of use and accurately finds non-coding RNAs with a low false discovery rate. Availability FindNonCoding is implemented within the DECIPHER package (v2.19.3) for R (v4.1) available from Bioconductor. Pre-trained models of common non-coding RNA families are included for bacteria, archaea, and eukarya. Supplementary information Supplementary data are available at Bioinformatics online.

scholarly journals Specific chromatin changes mark lateral organ founder cells in the Arabidopsis inflorescence meristem

2019 ◽  
Vol 70 (15) ◽  
pp. 3867-3879 ◽  
Author(s):  
Anneke Frerichs ◽  
Julia Engelhorn ◽  
Janine Altmüller ◽  
Jose Gutierrez-Marcos ◽  
Wolfgang Werr
Keyword(s):  
Dna Sequences ◽  
High Throughput Sequencing ◽  
Gene Activation ◽  
Regulatory Elements ◽  
Inflorescence Meristem ◽  
Genome Wide ◽  
A Genome ◽  
Hypersensitive Sites ◽  
Lateral Organ ◽  
Founder Cells

Abstract Fluorescence-activated cell sorting (FACS) and assay for transposase-accessible chromatin with high-throughput sequencing (ATAC-seq) were combined to analyse the chromatin state of lateral organ founder cells (LOFCs) in the peripheral zone of the Arabidopsis apetala1-1 cauliflower-1 double mutant inflorescence meristem. On a genome-wide level, we observed a striking correlation between transposase hypersensitive sites (THSs) detected by ATAC-seq and DNase I hypersensitive sites (DHSs). The mostly expanded DHSs were often substructured into several individual THSs, which correlated with phylogenetically conserved DNA sequences or enhancer elements. Comparing chromatin accessibility with available RNA-seq data, THS change configuration was reflected by gene activation or repression and chromatin regions acquired or lost transposase accessibility in direct correlation with gene expression levels in LOFCs. This was most pronounced immediately upstream of the transcription start, where genome-wide THSs were abundant in a complementary pattern to established H3K4me3 activation or H3K27me3 repression marks. At this resolution, the combined application of FACS/ATAC-seq is widely applicable to detect chromatin changes during cell-type specification and facilitates the detection of regulatory elements in plant promoters.

scholarly journals Integration of viral transcriptome sequencing with structure and sequence motifs predicts novel regulatory elements in SARS-CoV-2

2020 ◽  
Author(s):  
Brian J. Cox
Keyword(s):  
Regulatory Elements ◽  
Viral Gene ◽  
Template Switching ◽  
Sequence Motif ◽  
Sequence Motifs ◽  
Human Pathogens ◽  
Sequencing Data ◽  
Stem Loop ◽  
Conserved Sequence ◽  
Splice Junctions

SummaryIn the last twenty years, three separate coronaviruses have left their typical animal hosts and became human pathogens. An area of research interest is coronavirus transcription regulation that uses an RNA-RNA mediated template-switching mechanism. It is not known how different transcriptional stoichiometries of each viral gene are generated. Analysis of SARS-CoV-2 RNA sequencing data from whole RNA transcriptomes identified TRS dependent and independent transcripts. Integration of transcripts and 5’-UTR sequence motifs identified that the pentaloop and the stem-loop 3 were also located upstream of spliced genes. TRS independent transcripts were detected as likely non-polyadenylated. Additionally, a novel conserved sequence motif was discovered at either end of the TRS independent splice junctions. While similar both SARS viruses generated similar TRS independent transcripts they were more abundant in SARS-CoV-2. TRS independent gene regulation requires investigation to determine its relationship to viral pathogenicity.

scholarly journals Integration of methylation QTL and enhancer–target gene maps with schizophrenia GWAS summary results identifies novel genes

2019 ◽  
Vol 35 (19) ◽  
pp. 3576-3583 ◽  
Author(s):  
Chong Wu ◽  
Wei Pan
Keyword(s):  
Quantitative Trait ◽  
Large Scale ◽  
Target Gene ◽  
Association Studies ◽  
Regulatory Elements ◽  
Supplementary Information ◽  
Genome Wide Association Studies ◽  
Novel Genes ◽  
Gene Pairs ◽  
Genome Wide

Abstract Motivation Most trait-associated genetic variants identified in genome-wide association studies (GWASs) are located in non-coding regions of the genome and thought to act through their regulatory roles. Results To account for enriched association signals in DNA regulatory elements, we propose a novel and general gene-based association testing strategy that integrates enhancer-target gene pairs and methylation quantitative trait locus data with GWAS summary results; it aims to both boost statistical power for new discoveries and enhance mechanistic interpretability of any new discovery. By reanalyzing two large-scale schizophrenia GWAS summary datasets, we demonstrate that the proposed method could identify some significant and novel genes (containing no genome-wide significant SNPs nearby) that would have been missed by other competing approaches, including the standard and some integrative gene-based association methods, such as one incorporating enhancer-target gene pairs and one integrating expression quantitative trait loci. Availability and implementation Software: wuchong.org/egmethyl.html Supplementary information Supplementary data are available at Bioinformatics online.

i6mA-Pred: identifying DNA N6-methyladenine sites in the rice genome

2019 ◽  
Vol 35 (16) ◽  
pp. 2796-2800 ◽  
Author(s):  
Wei Chen ◽  
Hao Lv ◽  
Fulei Nie ◽  
Hao Lin
Keyword(s):  
Dna Sequences ◽  
Rice Genome ◽  
Chemical Properties ◽  
Computational Method ◽  
Accurate Identification ◽  
Feature Selection Technique ◽  
Jackknife Test ◽  
Genome Wide ◽  
Wide Range ◽  
User Friendly

Abstract Motivation DNA N6-methyladenine (6mA) is associated with a wide range of biological processes. Since the distribution of 6mA site in the genome is non-random, accurate identification of 6mA sites is crucial for understanding its biological functions. Although experimental methods have been proposed for this regard, they are still cost-ineffective for detecting 6mA site in genome-wide scope. Therefore, it is desirable to develop computational methods to facilitate the identification of 6mA site. Results In this study, a computational method called i6mA-Pred was developed to identify 6mA sites in the rice genome, in which the optimal nucleotide chemical properties obtained by the using feature selection technique were used to encode the DNA sequences. It was observed that the i6mA-Pred yielded an accuracy of 83.13% in the jackknife test. Meanwhile, the performance of i6mA-Pred was also superior to other methods. Availability and implementation A user-friendly web-server, i6mA-Pred is freely accessible at http://lin-group.cn/server/i6mA-Pred.

scholarly journals HUGIn: Hi-C Unifying Genomic Interrogator

2017 ◽  
Author(s):  
Joshua S. Martin ◽  
Zheng Xu ◽  
Alex P. Reiner ◽  
Karen L. Mohlke ◽  
Patrick Sullivan ◽  
...  
Keyword(s):  
High Throughput ◽  
Regulatory Elements ◽  
Supplementary Information ◽  
Chromatin Conformation ◽  
Organization Studies ◽  
Web Browser ◽  
Functional Roles ◽  
Gwas Snps ◽  
Genome Wide ◽  
Coding Variants

AbstractMotivationHigh throughput chromatin conformation capture (3C) technologies, such as Hi-C and ChlA-PET, have the potential to elucidate the functional roles of non-coding variants. However, most of published genome-wide unbiased chromatin organization studies have used cultured cell lines, limiting their generalizability.ResultsWe developed a web browser, HUGIn, to visualize Hi-C data generated from 21 human primary tissues and cell liens. HUGIn enables assessment of chromatin contacts both constitutive across and specific to tissue(s) and/or cell line(s) at any genomic loci, including GWAS SNPs, eQTLs and cis-regulatory elements, facilitating the understanding of both GWAS and eQTLs results and functional genomics data.AvailabilityHUGIn is available at http://yunliweb.its.unc.edu/[email protected] and [email protected] information:

scholarly journals Genome-wide cis-decoding for expression designing in tomato using cistrome data and explainable deep learning

2021 ◽  
Author(s):  
Takashi Akagi ◽  
Kanae Masuda ◽  
Eriko Kuwada ◽  
Kouki Takeshita ◽  
Taiji Kawakatsu ◽  
...  
Keyword(s):  
Expression Pattern ◽  
Dna Sequences ◽  
Regulatory Networks ◽  
Tomato Fruit ◽  
Expression Patterns ◽  
Regulatory Elements ◽  
Tomato Fruits ◽  
Genome Wide ◽  
Tomato Fruit Ripening ◽  
Single Cell Type

In the evolutionary paths of plants, variations of the cis-regulatory elements (CREs) resulting in expression diversification have played a central role in driving the establishment of lineage-specific traits. However, it is difficult to predict expression behaviors from the CRE patterns to properly harness them, mainly because the biological processes are complex. In this study, we used cistrome datasets and explainable convolutional neural network (CNN) frameworks to predict genome-wide expression patterns in tomato fruits from the DNA sequences in gene regulatory regions. By fixing the effects of trans-elements using single cell-type spatiotemporal transcriptome data for the response variables, we developed a prediction model of a key expression pattern for the initiation of tomato fruit ripening. Feature visualization of the CNNs identified nucleotide residues critical to the objective expression pattern in each gene and their effects, were validated experimentally in ripening tomato fruits. This cis-decoding framework will not only contribute to understanding the regulatory networks derived from CREs and transcription factor interactions, but also provide a flexible way of designing alleles with optimized expression.

scholarly journals ReFeaFi: Genome-wide prediction of regulatory elements driving transcription initiation

2021 ◽  
Author(s):  
Ramzan Umarov ◽  
Yu Li ◽  
Takahiro Arakawa ◽  
Satoshi Takizawa ◽  
Xin Gao ◽  
...  
Keyword(s):  
Gene Expression ◽  
Transcription Initiation ◽  
Expression Patterns ◽  
Regulatory Elements ◽  
Accurate Identification ◽  
Regulatory Regions ◽  
Genome Wide ◽  
Wide Scale ◽  
Genome Level ◽  
Level Performance

Regulatory elements control gene expression through transcription initiation (promoters) and by enhancing transcription at distant regions (enhancers). Accurate identification of regulatory elements is fundamental for annotating genomes and understanding gene expression patterns. While there are many attempts to develop computational promoter and enhancer identification methods, reliable tools to analyze long genomic sequences are still lacking. Prediction methods often perform poorly on the genome-wide scale because the number of negatives is much higher than that in the training sets. To address this issue, we propose a dynamic negative set updating scheme with a two-model approach, using one model for scanning the genome and the other one for testing candidate positions. The developed method achieves good genome-level performance and maintains robust performance when applied to other species, without re-training. Moreover, the unannotated predicted regulatory regions made on the human genome are enriched for disease-associated variants, suggesting them to be potentially true regulatory elements rather than false positives. We validated high scoring "false positive" predictions using reporter assay and all tested candidates were successfully validated, demonstrating the ability of our method to discover novel human regulatory regions.

scholarly journals Modelling the transcription factor DNA-binding affinity using genome-wide ChIP-based data

2016 ◽  
Author(s):  
Monther Alhamdoosh ◽  
Dianhui Wang
Keyword(s):  
Dna Binding ◽  
Binding Affinity ◽  
Dna Sequences ◽  
Binding Sites ◽  
Learning Algorithm ◽  
Regulatory Elements ◽  
Machine Learning Algorithms ◽  
Biological Knowledge ◽  
Genome Wide ◽  
Dna Binding Affinity

Understanding protein-DNA binding affinity is still a mystery for many transcription factors (TFs). Although several approaches have been proposed in the literature to model the DNA-binding specificity of TFs, they still have some limitations. Most of the methods require a cut-off threshold in order to classify a K-mer as a binding site (BS) and finding such a threshold is usually done by handcraft rather than a science. Some other approaches use a prior knowledge on the biological context of regulatory elements in the genome along with machine learning algorithms to build classifier models for TFBSs. Noticeably, these methods deliberately select the training and testing datasets so that they are very separable. Hence, the current methods do not actually capture the TF-DNA binding relationship. In this paper, we present a threshold-free framework based on a novel ensemble learning algorithm in order to locate TFBSs in DNA sequences. Our proposed approach creates TF-specific classifier models using genome-wide DNA-binding experiments and a prior biological knowledge on DNA sequences and TF binding preferences. Systematic background filtering algorithms are utilized to remove non-functional K-mers from training and testing datasets. To reduce the complexity of classifier models, a fast feature selection algorithm is employed. Finally, the created classifier models are used to scan new DNA sequences and identify potential binding sites. The analysis results show that our proposed approach is able to identify novel binding sites in the Saccharomyces cerevisiae [email protected], [email protected]://homepage.cs.latrobe.edu.au/dwang/DNNESCANweb

Sign in / Sign up

Export Citation Format

Share Document