skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


Title: Pangenome-Informed Language Models for Synthetic Genome Sequence Generation
Abstract Language Models (LM) have been extensively utilized for learning DNA sequence patterns and generating synthetic sequences. In this paper, we present a novel approach for the generation of synthetic DNA data using pangenomes in combination with LM. We introduce three innovative pangenome-based tokenization schemes, including two that can decouple from private data, while enhance long DNA sequence generation. Our experimental results demonstrate the superiority of pangenome-based tokenization over classical methods in generating high-utility synthetic DNA sequences, highlighting a promising direction for the public sharing of genomic datasets.  more » « less
Award ID(s):
2118743
PAR ID:
10630577
Author(s) / Creator(s):
; ; ; ; ; ;
Publisher / Repository:
bioRxiv
Date Published:
Format(s):
Medium: X
Institution:
bioRxiv
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; ; (, Bioinformatics Advances)
    Lengauer, Thomas (Ed.)
    Abstract SummaryThis article presents the Ensemble Nucleotide Byte-level Encoder-Decoder (ENBED) foundation model, analyzing DNA sequences at byte-level precision with an encoder–decoder Transformer architecture. ENBED uses a subquadratic implementation of attention to develop an efficient model capable of sequence-to-sequence transformations, generalizing previous genomic models with encoder-only or decoder-only architectures. We use Masked Language Modeling to pretrain the foundation model using reference genome sequences and apply it in the following downstream tasks: (i) identification of enhancers, promotors, and splice sites, (ii) recognition of sequences containing base call mismatches and insertion/deletion errors, an advantage over tokenization schemes involving multiple base pairs, which lose the ability to analyze with byte-level precision, (iii) identification of biological function annotations of genomic sequences, and (iv) generating mutations of the Influenza virus using the encoder–decoder architecture and validating them against real-world observations. In each of these tasks, we demonstrate significant improvement as compared to the existing state-of-the-art results. Availability and implementationThe source code used to develop and fine-tune the foundation model has been released on Github (https://github.itap.purdue.edu/Clan-labs/ENBED). 
    more » « less
  2. ; (, Bioinformatics)
    Alkan, Can (Ed.)
    Abstract Motivation Pangenome variation graphs model the mutual alignment of collections of DNA sequences. A set of pairwise alignments implies a variation graph, but there are no scalable methods to generate such a graph from these alignments. Existing related approaches depend on a single reference, a specific ordering of genomes or a de Bruijn model based on a fixed k-mer length. A scalable, self-contained method to build pangenome graphs without such limitations would be a key step in pangenome construction and manipulation pipelines. Results We design the seqwish algorithm, which builds a variation graph from a set of sequences and alignments between them. We first transform the alignment set into an implicit interval tree. To build up the variation graph, we query this tree-based representation of the alignments to reduce transitive matches into single DNA segments in a sequence graph. By recording the mapping from input sequence to output graph, we can trace the original paths through this graph, yielding a pangenome variation graph. We present an implementation that operates in external memory, using disk-backed data structures and lock-free parallel methods to drive the core graph induction step. We demonstrate that our method scales to very large graph induction problems by applying it to build pangenome graphs for several species. Availability and implementation seqwish is published as free software under the MIT open source license. Source code and documentation are available at https://github.com/ekg/seqwish. seqwish can be installed via Bioconda https://bioconda.github.io/recipes/seqwish/README.html or GNU Guix https://github.com/ekg/guix-genomics/blob/master/seqwish.scm. 
    more » « less
  3. ; ; ; (, Molecular Ecology Resources)
    Abstract Study of repetitive DNA elements in model organisms highlights the role of repetitive elements (REs) in many processes that drive genome evolution and phenotypic change. Because REs are much more dynamic than single‐copy DNA, repetitive sequences can reveal signals of evolutionary history over short time scales that may not be evident in sequences from slower‐evolving genomic regions. Many tools for studying REs are directed toward organisms with existing genomic resources, including genome assemblies and repeat libraries. However, signals in repeat variation may prove especially valuable in disentangling evolutionary histories in diverse non‐model groups, for which genomic resources are limited. Here, we introduce RepeatProfiler, a tool for generating, visualizing, and comparing repetitive element DNA profiles from low‐coverage, short‐read sequence data. RepeatProfiler automates the generation and visualization of RE coverage depth profiles (RE profiles) and allows for statistical comparison of profile shape across samples. In addition, RepeatProfiler facilitates comparison of profiles by extracting signal from sequence variants across profiles which can then be analysed as molecular morphological characters using phylogenetic analysis. We validate RepeatProfiler with data sets from ground beetles (Bembidion), flies (Drosophila), and tomatoes (Solanum). We highlight the potential of RE profiles as a high‐resolution data source for studies in species delimitation, comparative genomics, and repeat biology. 
    more » « less
  4. ; ; ; ; ; ; (, 2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW))
    Synthetic data is highly useful for training machine learning systems performing image-based 3D reconstruction, as synthetic data has applications in both extending existing generalizable datasets and being tailored to train neural networks for specific learning tasks of interest. In this paper, we introduce and utilize a synthetic data generation suite capable of generating data given existing 3D scene models as input. Specifically, we use our tool to generate image sequences for use with Multi-View Stereo (MVS), moving a camera through the virtual space according to user-chosen camera parameters. We evaluate how the given camera parameters and type of 3D environment affect how applicable the generated image sequences are to the MVS task using five pre-trained neural networks on image sequences generated from three different 3D scene datasets. We obtain generated predictions for each combination of parameter value and input image sequence, using standard error metrics to analyze the differences in depth predictions on image sequences across 3D datasets, parameters, and networks. Among other results, we find that camera height and vertical camera viewing angle are the parameters that cause the most variation in depth prediction errors on these image sequences. 
    more » « less
  5. ; ; ; ; ; ; ; (, Plant Biotechnology Journal)
    Summary Advancement of DNA‐synthesis technologies has greatly facilitated the development of synthetic biology tools. However, high‐complexity DNA sequences containing tandems of short repeats are still notoriously difficult to produce synthetically, with commercial DNA synthesis companies usually rejecting orders that exceed specific sequence complexity thresholds. To overcome this limitation, we developed a simple, single‐tube reaction method that enables the generation of DNA sequences containing multiple repetitive elements. Our strategy involves commercial synthesis and PCR amplification of padded sequences that contain the repeats of interest, along with random intervening sequence stuffers that include type IIS restriction enzyme sites. GoldenBraid molecular cloning technology is then employed to remove the stuffers, rejoin the repeats together in a predefined order, and subclone the tandem(s) in a vector using a single‐tube digestion–ligation reaction. In our hands, this new approach is much simpler, more versatile and efficient than previously developed solutions to this problem. As a proof of concept, two different phytohormone‐responsive, synthetic, repetitive proximal promoters were generated and testedin plantain the context of transcriptional reporters. Analysis of transgenic lines carrying the synthetic ethylene‐responsive promoter10x2EBS‐S10fused to theGUSreporter gene uncovered several developmentally regulated ethylene response maxima, indicating the utility of this reporter for monitoring the involvement of ethylene in a variety of physiologically relevant processes. These encouraging results suggest that this reporter system can be leveraged to investigate the ethylene response to biotic and abiotic factors with high spatial and temporal resolution. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email