BioSeq-BLM: a platform for analyzing DNA, RNA and protein sequences based on biological language models.
Ontology highlight
ABSTRACT:
SUBMITTER: Li HL
PROVIDER: S-EPMC8682797 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Ground-breaking progress has been made in structure prediction of biomolecular assemblies, including the recent breakthrough of AlphaFold 3. However, it remains challenging for AlphaFold 3 and other state-of-the-art deep learning-based methods to accurately predict protein-RNA complex structures, in part due to the limited availability of evolutionary and structural information related to protein-RNA interactions that are used as inputs to the existing approaches. Here, we introduce ProRNA3D-single, a new deep-learning framework for protein-RNA complex structure prediction with only single-sequence input. Using a novel geometric attention-enabled pairing of biological language models of protein and RNA, a previously unexplored avenue, ProRNA3D-single enables the prediction of interatomic protein-RNA interaction maps, which are then transformed into multi-scale geometric restraints for modeling 3D structures of protein-RNA complexes via geometry optimization. Benchmark tests show that ProRNA3D-single convincingly outperforms current state-of-the-art methods including AlphaFold 3, particularly when evolutionary information is limited; and exhibits remarkable robustness and performance resilience by attaining better accuracy with only single-sequence input than what most methods can achieve even with explicit evolutionary information. Freely available at https://github.com/Bhattacharya-Lab/ProRNA3D-single, ProRNA3D-single should be broadly useful for modeling 3D structures of protein-RNA complexes at scale, regardless of the availability of evolutionary information.
| S-EPMC11291176 | biostudies-literature
Project description:Deep-learning language models have shown promise in various biotechnological applications, including protein design and engineering. Here we describe ProGen, a language model that can generate protein sequences with a predictable function across large protein families, akin to generating grammatically and semantically correct natural language sentences on diverse topics. The model was trained on 280 million protein sequences from >19,000 families and is augmented with control tags specifying protein properties. ProGen can be further fine-tuned to curated sequences and tags to improve controllable generation performance of proteins from families with sufficient homologous samples. Artificial proteins fine-tuned to five distinct lysozyme families showed similar catalytic efficiencies as natural lysozymes, with sequence identity to natural proteins as low as 31.4%. ProGen is readily adapted to diverse protein families, as we demonstrate with chorismate mutase and malate dehydrogenase.
| S-EPMC10400306 | biostudies-literature
Project description:Genome instability often arises at common fragile sites (CFSs) leading to cancer-associated chromosomal rearrangements. However, the underlying mechanisms of how CFS protection is achieved is not well understood. We demonstrate that BLM plays an important role in the maintenance of genome stability of structure-forming AT-rich sequences derived from CFSs (CFS-AT). BLM deficiency leads to increased DSB formation and hyper mitotic recombination at CFS-AT and induces instability of the plasmids containing CFS-AT. We further showed that BLM is required for suppression of CFS breakage upon oncogene expression. Both helicase activity and ATR-mediated phosphorylation of BLM are important for preventing genetic instability at CFS-AT sequences. Furthermore, the role of BLM in protecting CFS-AT is not epistatic to that of FANCM, a translocase that is involved in preserving CFS stability. Loss of BLM helicase activity leads to drastic decrease of cell viability in FANCM deficient cells. We propose that BLM and FANCM utilize different mechanisms to remove DNA secondary structures forming at CFS-AT on replication forks, thereby preventing DSB formation and maintaining CFS stability.
| S-EPMC6289451 | biostudies-literature
Project description:Genomic studies have become noncoding RNA (ncRNA) centric after the study of different genomes provided enormous information on ncRNA over the past decades. The function of ncRNA is decided by its secondary structure, and across organisms, the secondary structure is more conserved than the sequence itself. In this study, the optimal secondary structure or the minimum free energy (MFE) structure of ncRNA was found based on the thermodynamic nearest neighbor model. MFE of over 2600 ncRNA sequences was analyzed in view of its signal properties. Mathematical models linking MFE to the signal properties were found for each of the four classes of ncRNA analyzed. MFE values computed with the proposed models were in concordance with those obtained with the standard web servers. A total of 95% of the sequences analyzed had deviation of MFE values within ±15% relative to those obtained from standard web servers.
| S-EPMC5029481 | biostudies-literature
Project description:The absence of functional BLM DNA helicase, a member of the RecQ family of helicases, is responsible for the rare human disorder Bloom Syndrome, which results in developmental abnormalities, DNA repair defects, genomic instability, and a predisposition to cancer. In Drosophila melanogaster, the orthologous Blm protein is essential during early development when the embryo is under the control of maternal gene products. We show that lack of functional maternal Blm during the syncytial cell cycles of Drosophila embryonic development results in severe nuclear defects and lethality. Amongst the small fraction of embryos from Blm mutant mothers that survive to adulthood, a prominent sex-bias favors the class that inherits less repetitive DNA content, which serves as an endogenous source of replication stress. This selection against repetitive DNA content reflects a role for Blm in facilitating replication through repetitive sequences during the rapid S-phases of syncytial cell cycles. During these syncytial cycles, Blm is not required for complex DNA double-strand break repair; however, the progeny sex-bias resulting from the absence of maternal Blm is exacerbated by repetitive DNA sequences and by the slowing of replication fork progression, suggesting that the essential role for Blm during this stage is to manage replication fork stress brought about by impediments to fork progression. Additionally, our data suggest that Blm is only required to manage this replication stress during embryonic development, and likely only during the early, rapid syncytial cell cycles, and not at later developmental stages. These results provide novel insights into Blm function throughout development.
| S-EPMC8733427 | biostudies-literature
Project description:Measuring allele-specific RNA expression provides valuable insights into cis-acting genetic and epigenetic regulation of gene expression. Widespread adoption of high-throughput sequencing technologies for studying RNA expression (RNA-Seq) permits measurement of allelic RNA expression imbalance (AEI) at heterozygous single nucleotide polymorphisms (SNPs) across the entire transcriptome, and this approach has become especially popular with the emergence of large databases, such as GTEx. However, the existing binomial-type methods used to model allelic expression from RNA-seq assume a strong negative correlation between reference and variant allele reads, which may not be reasonable biologically.Here we propose a new strategy for AEI analysis using RNA-seq data. Under the null hypothesis of no AEI, a group of SNPs (possibly across multiple genes) is considered comparable if their respective total sums of the allelic reads are of similar magnitude. Within each group of "comparable" SNPs, we identify SNPs with AEI signal by fitting a mixture of folded Skellam distributions to the absolute values of read differences. By applying this methodology to RNA-Seq data from human autopsy brain tissues, we identified numerous instances of moderate to strong imbalanced allelic RNA expression at heterozygous SNPs. Findings with SLC1A3 mRNA exhibiting known expression differences are discussed as examples.The folded Skellam mixture model searches for SNPs with significant difference between reference and variant allele reads (adjusted for different library sizes), using information from a group of "comparable" SNPs across multiple genes. This model is particularly suitable for performing AEI analysis on genes with few heterozygous SNPs available from RNA-seq, and it can fit over-dispersed read counts without specifying the direction of the correlation between reference and variant alleles.
| S-EPMC4521363 | biostudies-literature
Project description:After years of development, the complexity of the biological sequence alignment algorithm is gradually increasing, and the lack of high abstract level domain research leads to the complexity of its algorithm development and improvement. By applying the idea of software components to the design and development of algorithms, the development efficiency and reliability of biological sequence alignment algorithms can be effectively improved. The component assembly platform applies related assembly technology, which simplifies the operation difficulty of component assembly and facilitates the maintenance and optimization of the algorithm. At the same time, a friendly visual interface is used to intuitively complete the assembly of algorithm components, and an executable sequence alignment algorithm program is obtained, which can directly carry out alignment computing.
| S-EPMC7859483 | biostudies-literature
Project description:Predicting which proteins interact together from amino acid sequences is an important task. We develop a method to pair interacting protein sequences which leverages the power of protein language models trained on multiple sequence alignments (MSAs), such as MSA Transformer and the EvoFormer module of AlphaFold. We formulate the problem of pairing interacting partners among the paralogs of two protein families in a differentiable way. We introduce a method called Differentiable Pairing using Alignment-based Language Models (DiffPALM) that solves it by exploiting the ability of MSA Transformer to fill in masked amino acids in multiple sequence alignments using the surrounding context. MSA Transformer encodes coevolution between functionally or structurally coupled amino acids within protein chains. It also captures inter-chain coevolution, despite being trained on single-chain data. Relying on MSA Transformer without fine-tuning, DiffPALM outperforms existing coevolution-based pairing methods on difficult benchmarks of shallow multiple sequence alignments extracted from ubiquitous prokaryotic protein datasets. It also outperforms an alternative method based on a state-of-the-art protein language model trained on single sequences. Paired alignments of interacting protein sequences are a crucial ingredient of supervised deep learning methods to predict the three-dimensional structure of protein complexes. Starting from sequences paired by DiffPALM substantially improves the structure prediction of some eukaryotic protein complexes by AlphaFold-Multimer. It also achieves competitive performance with using orthology-based pairing.
| S-EPMC11228504 | biostudies-literature
Project description:The vast amount of RNA-seq data deposited in Gene Expression Omnibus (GEO) and Sequence Read Archive (SRA) is still a grossly underutilized resource for biomedical research. To remove technical roadblocks for reusing these data, we have developed a web-application GREIN (GEO RNA-seq Experiments Interactive Navigator) which provides user-friendly interfaces to manipulate and analyze GEO RNA-seq data. GREIN is powered by the back-end computational pipeline for uniform processing of RNA-seq data and the large number (>6,500) of already processed datasets. The front-end user interfaces provide a wealth of user-analytics options including sub-setting and downloading processed data, interactive visualization, statistical power analyses, construction of differential gene expression signatures and their comprehensive functional characterization, and connectivity analysis with LINCS L1000 data. The combination of the massive amount of back-end data and front-end analytics options driven by user-friendly interfaces makes GREIN a unique open-source resource for re-using GEO RNA-seq data. GREIN is accessible at: https://shiny.ilincs.org/grein , the source code at: https://github.com/uc-bd2k/grein , and the Docker container at: https://hub.docker.com/r/ucbd2k/grein .
| S-EPMC6527554 | biostudies-literature
Project description:Topological entropy is one of the most difficult entropies to be used to analyze the DNA sequences, due to the finite sample and high-dimensionality problems. In order to overcome these problems, a generalized topological entropy is introduced. The relationship between the topological entropy and the generalized topological entropy is compared, which shows the topological entropy is a special case of the generalized entropy. As an application the generalized topological entropy in introns, exons and promoter regions was computed, respectively. The results indicate that the entropy of introns is higher than that of exons, and the entropy of the exons is higher than that of the promoter regions for each chromosome, which suggest that DNA sequence of the promoter regions is more regular than the exons and introns.
| S-EPMC3922877 | biostudies-literature