Bidirectional de novo peptide sequencing using a transformer model
Ontology highlight
ABSTRACT:
SUBMITTER: Lee S
PROVIDER: S-EPMC10901305 | biostudies-literature | 2024 Feb
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:The primary objective of analyzing the data obtained in a mass spectrometry-based proteomic experiment is peptide and protein identification, or correct assignment of the tandem mass spectrum to one amino acid sequence. Comparison of empirical fragment spectra with the theoretical predicted one or matching with the collected spectra library are commonly accepted strategies of proteins identification and defining of their amino acid sequences. Although these approaches are widely used and are appreciably efficient for the well-characterized model organisms or measured proteins, they cannot detect novel peptide sequences that have not been previously annotated or are rare. This study presents PowerNovo tool for de novo sequencing of proteins using tandem mass spectra acquired in a variety of types of mass analyzers and different fragmentation techniques. PowerNovo involves an ensemble of models for peptide sequencing: model for detecting regularities in tandem mass spectra, precursors, and fragment ions and a natural language processing model, which has a function of peptide sequence quality assessment and helps with reconstruction of noisy sequences. The results of testing showed that the performance of PowerNovo is comparable and even better than widely utilized PointNovo, DeepNovo, Casanovo, and Novor packages. Also, PowerNovo provides complete cycle of processing (pipeline) of mass spectrometry data and, along with predicting the peptide sequence, involves the peptide assembly and protein inference blocks.
| S-EPMC11217302 | biostudies-literature
Project description:De novo peptide sequencing directly identifies peptides from mass spectrometry data, playing a critical role in discovering novel proteins and analyzing complex biological samples without reliance on existing databases. To address challenges in both speed and accuracy, a transformer-based model, TSARseqNovo, incorporates two key innovations: a Semi-Autoregressive decoder for parallel prediction of multiple amino acids and a Masking Refinement decoder for refining low-confidence predictions. These features significantly enhance sequencing efficiency and accuracy. Evaluations on the Nine-Species, Aggregated, and Glycoproteomic datasets, demonstrate that TSARseqNovo outperforms state-of-the-art models, including CasaNovo, NovoB, InstaNovo + , and π-HelixNovo. Specifically, TSARseqNovo achieves up to a 2-fold speed increase over CasaNovo and π-HelixNovo, and approximately 10-fold over NovoB and InstaNovo + , while also showing substantial improvements in peptide prediction precision, especially for long peptides. These advancements position TSARseqNovo as a powerful tool for accelerating high-throughput proteomics research and addressing increasingly complex biological questions.
| S-EPMC11825679 | biostudies-literature
Project description:De novo peptide sequencing, which does not rely on a comprehensive target sequence database, provides us with a way to identify novel peptides from tandem mass spectra. However, current de novo sequencing algorithms suffer from low accuracy and coverage, which hinders their application in proteomics. In this paper, we present PepNet, a fully convolutional neural network for high accuracy de novo peptide sequencing. PepNet takes an MS/MS spectrum (represented as a high-dimensional vector) as input, and outputs the optimal peptide sequence along with its confidence score. The PepNet model is trained using a total of 3 million high-energy collisional dissociation MS/MS spectra from multiple human peptide spectral libraries. Evaluation results show that PepNet significantly outperforms current best-performing de novo sequencing algorithms (e.g. PointNovo and DeepNovo) in both peptide-level accuracy and positional-level accuracy. PepNet can sequence a large fraction of spectra that were not identified by database search engines, and thus could be used as a complementary tool to database search engines for peptide identification in proteomics. In addition, PepNet runs around 3x and 7x faster than PointNovo and DeepNovo on GPUs, respectively, thus being more suitable for the analysis of large-scale proteomics data.
| S-EPMC10693636 | biostudies-literature
Project description:Proliferation of drug-resistant diseases raises the challenge of searching for new, more efficient antibiotics. Currently, some of the most effective antibiotics (i.e., Vancomycin and Daptomycin) are cyclic peptides produced by non-ribosomal biosynthetic pathways. The isolation and sequencing of cyclic peptide antibiotics, unlike the same activity with linear peptides, is time-consuming and error-prone. The dominant technique for sequencing cyclic peptides is nuclear magnetic resonance (NMR)-based and requires large amounts (milligrams) of purified materials that, for most compounds, are not possible to obtain. Given these facts, there is a need for new tools to sequence cyclic non-ribosomal peptides (NRPs) using picograms of material. Since nearly all cyclic NRPs are produced along with related analogs, we develop a mass spectrometry approach for sequencing all related peptides at once (in contrast to the existing approach that analyzes individual peptides). Our results suggest that instead of attempting to isolate and NMR-sequence the most abundant compound, one should acquire spectra of many related compounds and sequence all of them simultaneously using tandem mass spectrometry. We illustrate applications of this approach by sequencing new variants of cyclic peptide antibiotics from Bacillus brevis, as well as sequencing a previously unknown family of cyclic NRPs produced by marine bacteria. Supplementary Material is available online at www.liebertonline.com/cmb.
| S-EPMC3216106 | biostudies-literature
Project description:De novo peptide sequencing from tandem MS data is the key technology in proteomics for the characterization of proteins, especially for new sequences, such as mAbs. In this study, we propose a deep neural network model, DeepNovo, for de novo peptide sequencing. DeepNovo architecture combines recent advances in convolutional neural networks and recurrent neural networks to learn features of tandem mass spectra, fragment ions, and sequence patterns of peptides. The networks are further integrated with local dynamic programming to solve the complex optimization task of de novo sequencing. We evaluated the method on a wide variety of species and found that DeepNovo considerably outperformed state of the art methods, achieving 7.7-22.9% higher accuracy at the amino acid level and 38.1-64.0% higher accuracy at the peptide level. We further used DeepNovo to automatically reconstruct the complete sequences of antibody light and heavy chains of mouse, achieving 97.5-100% coverage and 97.2-99.5% accuracy, without assisting databases. Moreover, DeepNovo is retrainable to adapt to any sources of data and provides a complete end-to-end training and prediction solution to the de novo sequencing problem. Not only does our study extend the deep learning revolution to a new field, but it also shows an innovative approach in solving optimization problems by using deep learning and dynamic programming.
| S-EPMC5547637 | biostudies-literature
Project description:De novo sequencing software has been widely used in proteomics to sequence new peptides from tandem mass spectrometry data. This study presents a new software tool, Novor, to greatly improve both the speed and accuracy of today's peptide de novo sequencing analyses. To improve the accuracy, Novor's scoring functions are based on two large decision trees built from a peptide spectral library with more than 300,000 spectra with machine learning. Important knowledge about peptide fragmentation is extracted automatically from the library and incorporated into the scoring functions. The decision tree model also enables efficient score calculation and contributes to the speed improvement. To further improve the speed, a two-stage algorithmic approach, namely dynamic programming and refinement, is used. The software program was also carefully optimized. On the testing datasets, Novor sequenced 7%-37% more correct residues than the state-of-the-art de novo sequencing tool, PEAKS, while being an order of magnitude faster. Novor can de novo sequence more than 300 MS/MS spectra per second on a laptop computer. The speed surpasses the acquisition speed of today's mass spectrometer and, therefore, opens a new possibility to de novo sequence in real time while the spectrometer is acquiring the spectral data. Graphical Abstract ᅟ.
| S-EPMC4604512 | biostudies-literature
Project description:MotivationDe novo peptide sequencing based on tandem mass spectrometry data is the key technology of shotgun proteomics for identifying peptides without any database and assembling unknown proteins. However, owing to the low ion coverage in tandem mass spectra, the order of certain consecutive amino acids cannot be determined if all of their supporting fragment ions are missing, which results in the low precision of de novo sequencing.ResultsIn order to solve this problem, we developed pNovo 3, which used a learning-to-rank framework to distinguish similar peptide candidates for each spectrum. Three metrics for measuring the similarity between each experimental spectrum and its corresponding theoretical spectrum were used as important features, in which the theoretical spectra can be precisely predicted by the pDeep algorithm using deep learning. On seven benchmark datasets from six diverse species, pNovo 3 recalled 29-102% more correct spectra, and the precision was 11-89% higher than three other state-of-the-art de novo sequencing algorithms. Furthermore, compared with the newly developed DeepNovo, which also used the deep learning approach, pNovo 3 still identified 21-50% more spectra on the nine datasets used in the study of DeepNovo. In summary, the deep learning and learning-to-rank techniques implemented in pNovo 3 significantly improve the precision of de novo sequencing, and such machine learning framework is worth extending to other related research fields to distinguish the similar sequences.Availability and implementationpNovo 3 can be freely downloaded from http://pfind.ict.ac.cn/software/pNovo/index.html.Supplementary informationSupplementary data are available at Bioinformatics online.
| S-EPMC6612832 | biostudies-literature
Project description:The impact of mixture spectra deconvolution on the performance of four popular de novo sequencing programs was tested using artificially constructed mixture spectra as well as experimental proteomics data. Mixture fragmentation spectra are recognized as a limitation in proteomics because they decrease the identification performance using database search engines. De novo sequencing approaches are expected to be even more sensitive to the reduction in mass spectrum quality resulting from peptide precursor co-isolation and thus prone to false identifications. The deconvolution approach matched complementary b-, y-ions to each precursor peptide mass, which allowed the creation of virtual spectra containing sequence specific fragment ions of each co-isolated peptide. Deconvolution processing resulted in equally efficient identification rates but increased the absolute number of correctly sequenced peptides. The improvement was in the range of 20-35% additional peptide identifications for a HeLa lysate sample. Some correct sequences were identified only using unprocessed spectra; however, the number of these was lower than those where improvement was obtained by mass spectral deconvolution. Tight candidate peptide score distribution and high sensitivity to small changes in the mass spectrum introduced by the employed deconvolution method could explain some of the missing peptide identifications.
| S-EPMC5297990 | biostudies-literature
Project description:MotivationMass spectrometry (MS) instruments and experimental protocols are rapidly advancing, but de novo peptide sequencing algorithms to analyze tandem mass (MS/MS) spectra are lagging behind. Although existing de novo sequencing tools perform well on certain types of spectra [e.g. Collision Induced Dissociation (CID) spectra of tryptic peptides], their performance often deteriorates on other types of spectra, such as Electron Transfer Dissociation (ETD), Higher-energy Collisional Dissociation (HCD) spectra or spectra of non-tryptic digests. Thus, rather than developing a new algorithm for each type of spectra, we develop a universal de novo sequencing algorithm called UniNovo that works well for all types of spectra or even for spectral pairs (e.g. CID/ETD spectral pairs). UniNovo uses an improved scoring function that captures the dependences between different ion types, where such dependencies are learned automatically using a modified offset frequency function.ResultsThe performance of UniNovo is compared with PepNovo+, PEAKS and pNovo using various types of spectra. The results show that the performance of UniNovo is superior to other tools for ETD spectra and superior or comparable with others for CID and HCD spectra. UniNovo also estimates the probability that each reported reconstruction is correct, using simple statistics that are readily obtained from a small training dataset. We demonstrate that the estimation is accurate for all tested types of spectra (including CID, HCD, ETD, CID/ETD and HCD/ETD spectra of trypsin, LysC or AspN digested peptides).AvailabilityUniNovo is implemented in JAVA and tested on Windows, Ubuntu and OS X machines. UniNovo is available at http://proteomics.ucsd.edu/Software/UniNovo.html along with the manual.
| S-EPMC3722526 | biostudies-literature
Project description:Accurately building three-dimensional (3D) atomic structures from 3D cryo-electron microscopy (cryo-EM) density maps is a crucial step in the cryo-EM-based determination of the structures of protein complexes. Despite improvements in the resolution of 3D cryo-EM density maps, the de novo conversion of density maps into 3D atomic structures for protein complexes that do not have accurate homologous or predicted structures to be used as templates remains a significant challenge. Here, we introduce Cryo2Struct, a fully automated ab initio cryo-EM structure modeling method that utilizes a 3D transformer to identify atoms and amino acid types in cryo-EM density maps first, and then employs a novel Hidden Markov Model (HMM) to connect predicted atoms to build backbone structures of proteins. Tested on a standard test dataset of 128 cryo-EM density maps with varying resolutions (2.1 - 5.6 °A) and different numbers of residues (730 - 8,416), Cryo2Struct built substantially more accurate and complete protein structural models than the widely used ab initio method - Phenix in terms of multiple evaluation metrics. Moreover, on a new test dataset of 500 recently released density maps with varying resolutions (1.9 - 4.0 °A) and different numbers of residues (234 - 8,828), it built more accurate models than on the standard dataset. And its performance is rather robust against the change of the resolution of density maps and the size of protein structures.
| S-EPMC10802328 | biostudies-literature