Toolkits and Libraries for Deep Learning.
Ontology highlight
ABSTRACT:
SUBMITTER: Erickson BJ
PROVIDER: S-EPMC5537091 | biostudies-other | 2017 Aug
REPOSITORIES: biostudies-other
Similar Datasets
Project description:Data-independent acquisition (DIA) is an emerging technology for quantitative proteomic analysis of large cohorts of samples. However, sample-specific spectral libraries built by data-dependent acquisition (DDA) experiments are required prior to DIA analysis, which is time-consuming and limits the identification/quantification by DIA to the peptides identified by DDA. Herein, we propose DeepDIA, a deep learning-based approach to generate in silico spectral libraries for DIA analysis. We demonstrate that the quality of in silico libraries predicted by instrument-specific models using DeepDIA is comparable to that of experimental libraries, and outperforms libraries generated by global models. With peptide detectability prediction, in silico libraries can be built directly from protein sequence databases. We further illustrate that DeepDIA can break through the limitation of DDA on peptide/protein detection, and enhance DIA analysis on human serum samples compared to the state-of-the-art protocol using a DDA library. We expect this work expanding the toolbox for DIA proteomics.
| S-EPMC6952453 | biostudies-literature
Project description:Motivation:Motif discovery in large biopolymer sequence datasets can be computationally demanding, presenting significant challenges for discovery in omics research. MEME, arguably one of the most popular motif discovery software, takes quadratic time with respect to dataset size, leading to excessively long runtimes for large datasets. Therefore, there is a demand for fast programs that can generate results of the same quality as MEME. Results:Here we describe YAMDA, a highly scalable motif discovery software package. It is built on Pytorch, a tensor computation deep learning library with strong GPU acceleration that is highly optimized for tensor operations that are also useful for motifs. YAMDA takes linear time to find motifs as accurately as MEME, completing in seconds or minutes, which translates to speedups over a thousandfold. Availability and implementation:YAMDA is freely available on Github (https://github.com/daquang/YAMDA). Supplementary information:Supplementary data are available at Bioinformatics online.
| S-EPMC6184538 | biostudies-literature
Project description:Chromatin immunoprecipitation followed by high-throughput (HTP) sequencing (ChIP-seq) is a powerful tool to establish protein-DNA interactions genome-wide. The primary limitation of its broad application at present is the often-limited access to sequencers. Here we report a protocol, Mab-seq, that generates genome-scale quality evaluations for nucleic acid libraries intended for deep-sequencing. We show how commercially available genomic microarrays can be used to maximize the efficiency of library creation and quickly generate reliable preliminary data on a chromosomal scale in advance of deep sequencing. We also exploit this technique to compare enriched regions identified using microarrays with those identified by sequencing, demonstrating that they agree on a core set of clearly identified enriched regions, while characterizing the additional enriched regions identifiable using HTP sequencing.
| S-EPMC2577887 | biostudies-literature
Project description:In this work, we present a deep learning based full-spectrum prediction method and demonstrate the merits of using full-spectrum predicted approaches for library searching. Our proposed model provides flexibility to accommodate post-translational modifications and fills the current gap for long peptide predictions.
2023-06-28 | PXD040721 | Pride
Project description:Data-independent acquisition (DIA) proteomics allows systematic and unbiased measurement of protein samples and enables fast quantitative analysis of large cohorts of samples. However, sample-specific spectral libraries are usually required prior to perform DIA experiments. The libraries are built by data-dependent acquisition (DDA) proteomic analysis on the same samples normally with pre-fractionation, which is time-consuming and limits the identification/quantification by DIA to the peptides identified by DDA. Herein, we propose DeepDIA, a deep learning based approach to generate in silico spectral libraries for DIA analysis. We benchmarked DeepDIA with HeLa and mixed proteome data sets, showing that the quality of in silico spectral libraries is comparable to that of experimental spectral libraries. We further demonstrated that DeepDIA can be performed on human cell lines and human serum without pre-knowledge of peptides lists by DDA experiments. Compared to the state-of-the-art protocol using DDA-based spectral library with high abundance protein depletion and pre-fractionation, the number of identified and quantified proteins from human serum was increased by >100% using DeepDIA. Accuracy of the identification results was validated using a standard mixture containing >800 stable isotope labelled reference peptides from >500 proteins in human plasma. We expect this work contributing to the studies of quantitative proteomics and especially blood proteomics, whereas expanding the toolbox for DIA proteomics.
2019-12-02 | PXD014108 | iProX
Project description:Domain recombination is a key principle in protein evolution and protein engineering, but inserting a donor domain into every position of a target protein is not easily experimentally accessible. Most contemporary domain insertion profiling approaches rely on DNA transposons, which are constrained by sequence bias. Here, we establish Saturated Programmable Insertion Engineering (SPINE), an unbiased, comprehensive, and targeted domain insertion library generation technique using oligo library synthesis and multi-step Golden Gate cloning. Through benchmarking to MuA transposon-mediated library generation on four ion channel genes, we demonstrate that SPINE-generated libraries are enriched for in-frame insertions, have drastically reduced sequence bias as well as near-complete and highly-redundant coverage. Unlike transposon-mediated domain insertion that was severely biased and sparse for some genes, SPINE generated high-quality libraries for all genes tested. Using the Inward Rectifier K+ channel Kir2.1, we validate the practical utility of SPINE by constructing and comparing domain insertion permissibility maps. SPINE is the first technology to enable saturated domain insertion profiling. SPINE could help explore the relationship between domain insertions and protein function, and how this relationship is shaped by evolutionary forces and can be engineered for biomedical applications.
| S-EPMC6954442 | biostudies-literature
Project description:MicroRNA (miRNA) and other types of small regulatory RNAs play a crucial role in the regulation of gene expression in eukaryotes. Several distinct classes of small regulatory RNAs have been discovered in recent years. To extend the repertoire of small regulatory RNAs characterized in chickens we used a deep sequencing approach developed by Solexa (now Illumina Inc.). We sequenced three small RNA libraries prepared from different developmental stages of the chicken embryo (days 5, 7, and 9) to produce over 9.5 million short sequence reads. We developed a bioinformatics pipeline to distinguish authentic mature miRNA sequences from other classes of small RNAs and short RNA fragments represented in the sequencing data. Using this approach we detected almost all of the previously known chicken miRNAs and their respective miRNA* sequences. In addition we discovered 449 putative new chicken miRNAs. Of these, 430 miRNAs appear to be specific to the avian lineage. Another 6 new miRNAs had evidence of evolutionary conservation in at least one vertebrate species outside of the bird lineage. The remaining 13 putative miRNAs appear to represent chicken orthologs of known vertebrate miRNAs. We discovered 39 additional putative miRNA candidates originating from miRNA generating intronic sequences known as mirtrons. Keywords: miRNA discovery, mirtrons, chicken embryo
2008-05-12 | GSE10686 | GEO
Project description:A large number of small RNAs unrelated to the soybean genome were identified after deep sequencing of soybean small RNA libraries. A metatranscriptomic analysis was carried out to identify the origin of these sequences. Comparative analyses of small interference RNAs (siRNAs) present in samples collected in open areas corresponding to soybean field plantations and samples from soybean cultivated in greenhouses under a controlled environment were made. Different pathogenic, symbiotic and free-living organisms were identified from samples of both growth systems. They included viruses, bacteria and different groups of fungi. This approach can be useful not only to identify potentially unknown pathogens and pests, but also to understand the relations that soybean plants establish with microorganisms that may affect, directly or indirectly, plant health and crop production.
| S-EPMC3392881 | biostudies-other
Project description:Even though ghost imaging (GI), an unconventional imaging method, has received increased attention by researchers during the last decades, imaging speed is still not satisfactory. Once the data-acquisition method and the system parameters are determined, only the processing method has the potential to accelerate image-processing significantly. However, both the basic correlation method and the compressed sensing algorithm, which are often used for ghost imaging, have their own problems. To overcome these challenges, a novel deep learning ghost imaging method is proposed in this paper. We modified the convolutional neural network that is commonly used in deep learning to fit the characteristics of ghost imaging. This modified network can be referred to as ghost imaging convolutional neural network. Our simulations and experiments confirm that, using this new method, a target image can be obtained faster and more accurate at low sampling rate compared with conventional GI method.
| S-EPMC5915577 | biostudies-other
Project description:To reduce experimental effort associated with directed protein evolution and to explore the sequence space encoded by mutating multiple positions simultaneously, we incorporate machine learning into the directed evolution workflow. Combinatorial sequence space can be quite expensive to sample experimentally, but machine-learning models trained on tested variants provide a fast method for testing sequence space computationally. We validated this approach on a large published empirical fitness landscape for human GB1 binding protein, demonstrating that machine learning-guided directed evolution finds variants with higher fitness than those found by other directed evolution approaches. We then provide an example application in evolving an enzyme to produce each of the two possible product enantiomers (i.e., stereodivergence) of a new-to-nature carbene Si-H insertion reaction. The approach predicted libraries enriched in functional enzymes and fixed seven mutations in two rounds of evolution to identify variants for selective catalysis with 93% and 79% ee (enantiomeric excess). By greatly increasing throughput with in silico modeling, machine learning enhances the quality and diversity of sequence solutions for a protein engineering problem.
| S-EPMC6500146 | biostudies-literature