Article
Pharmacology & Pharmacy
Sara Carillo, Angela Criscuolo, Florian Fuessl, Ken Cook, Jonathan Bones
Summary: This study combines intact protein mass spectrometry and the multi-attribute method to create an intact multi-attribute method (iMAM) for defining and quantifying biopharmaceutical critical quality attributes. The results show the potential of iMAM for implementation at different stages of the production pipeline.
EUROPEAN JOURNAL OF PHARMACEUTICS AND BIOPHARMACEUTICS
(2022)
Article
Geosciences, Multidisciplinary
Jingbao Zhu, Shanyou Li, Yongxiang Wei, Jindong Song
Summary: By combining recurrent neural network (RNN) and transfer learning, we propose a method to predict on-site PGA and PGV in China, which shows better performance compared to traditional methods. Furthermore, the predicted instrumental seismic intensity is consistent with the observed results, with minimal error.
JOURNAL OF ASIAN EARTH SCIENCES
(2023)
Article
Chemistry, Analytical
Max Reuschenbach, Felix Drees, Torsten C. Schmidt, Gerrit Renner
Summary: In this study, a novel algorithm called qBinning is proposed to construct extracted ion chromatograms (EICs) based on statistical principles, eliminating the need for user parameter settings. A scoring system (DQS(bin)) is also introduced to provide user feedback on the specific qualities of the generated EICs. This work is significant for understanding the behavior of non-target screening (NTS) data and enhancing the overall transparency in NTS results.
ANALYTICAL CHEMISTRY
(2023)
Article
Chemistry, Analytical
Jun Sang Yu, Louis-Felix Nothias, Mingxun Wang, Dong Hyun Kim, Pieter C. Dorrestein, Kyo Bin Kang, Hye Hyun Yoo
Summary: This study compared the performance of molecular networking with existing strategies for drug metabolite identification. The results showed that molecular networking exhibited comparable or superior performance in terms of the number of detected metabolites, false positives ratio, and the amount of time and effort required for human labor-based postprocessing.
ANALYTICAL CHEMISTRY
(2022)
Article
Biochemical Research Methods
Ruimin Wang, Miaoshan Lu, Shaowei An, Jinyin Wang, Changbin Yu
Summary: In this study, a novel deep learning-based model called 3D-MSNet was proposed for untargeted feature extraction in metabolomics research. Comparisons with nine popular software demonstrated that 3D-MSNet outperformed others in feature detection and quantification accuracy. Additionally, 3D-MSNet exhibited high feature extraction robustness and could be applied to profile MS data acquired with various high-resolution mass spectrometers.
Article
Chemistry, Analytical
Sean M. Colby, Christine H. Chang, Jessica L. Bade, Jamie R. Nunez, Madison R. Blumer, Daniel J. Orton, Kent J. Bloodsworth, Ernesto S. Nakayasu, Richard D. Smith, Yehia M. Ibrahim, Ryan S. Renslow, Thomas O. Metz
Summary: DEIMoS is a Python tool for high-dimensional mass spectrometry data analysis that provides features such as feature detection, feature alignment, and calibration. It operates on N-dimensional data and can improve detection sensitivity and data alignment confidence while reducing artifacts in tandem mass spectra. The demonstration with metabolomics data shows the advantages of using DEIMoS in each data processing step.
ANALYTICAL CHEMISTRY
(2022)
Article
Biochemical Research Methods
Biying Chen, Chenxi Wang, Zhifei Fu, Haiyang Yu, Erwei Liu, Xiumei Gao, Jie Li, Lifeng Han
Summary: In this study, an ensemble model named RT-Ensemble Pred was successfully built based on QSRRs to predict the retention time of different LC-MS systems. By using ensemble sampling, RT-Ensemble Pred could better utilize online datasets for retention time prediction. The results showed that RT-Ensemble Pred could predict metabolites not included in the database and metabolites from new LC-MS methods, and it can also be used for the prediction of other compounds.
JOURNAL OF CHROMATOGRAPHY A
(2023)
Article
Oncology
Magdalena Buszewska-Forajta, Pawel Pomastowski, Fernanda Monedeiro, Justyna Walczak-Skierska, Marcin Markuszewski, Marcin Matuszewski, Michal J. Markuszewski, Boguslaw Buszewski
Summary: Prostate cancer is a major cause of cancer death in men, often recognized late due to lack of fast diagnostic methods. This study used mass spectrometry to analyze lipid alterations in prostate tissue, aiming to develop a quick and sensitive diagnostic method and improve hormone therapies effectiveness. The results suggest that lipidomics could serve as an alternative diagnostic tool for prostate cancer diagnosis.
Article
Water Resources
Mistaya Langridge, Ed McBean, Hossein Bonakdari, Bahram Gharabaghi
Summary: A simplified empirical equation has been developed for predicting dynamic catchment response time, allowing for variability between storm sizes and catchment moisture conditions. This model has been translated to North America for the first time and has shown statistical success in both the UK and North America through selected study areas in Canada and the United States. Additionally, a runoff coefficient has been introduced to encompass historical catchment wetness, making it easier to incorporate antecedent moisture condition into predictions.
HYDROLOGICAL PROCESSES
(2021)
Article
Biochemical Research Methods
Kyle J. Juetten, Jennifer S. Brodbelt
Summary: MS-TAFI is a free Python-based program for analyzing and visualizing deconvoluted MS/MS data of intact proteins, offering a streamlined approach to data analysis. It also includes tools for native mass spectrometry experiments, allowing for the search of fragment ions with retained ligands and visualization of charge site locations.
JOURNAL OF PROTEOME RESEARCH
(2022)
Article
Chemistry, Analytical
Jiawen Ai, Weize Zhao, Quan Yu, Xiang Qian, Jianhua Zhou, Xinming Huo, Fei Tang
Summary: The mass spectrometer is an important tool for chemical analysis and detection, and the emergence of miniature mass spectrometers has provided new tools for field analysis and detection. In this study, a super-resolution U-net algorithm was proposed for ion trap mass spectrometry, which improved the resolution and detection capabilities of the instrument.
ANALYTICAL CHEMISTRY
(2023)
Article
Astronomy & Astrophysics
Lluis Mas-Ribas, Guochao Sun, Tzu-Ching Chang, Michael O. Gonzalez, Richard H. Mebane
Summary: We introduce LIMFAST, a seminumerical code for simulating high-redshift galaxy formation and cosmic reionization using multitracer line intensity mapping (LIM) signals. LIMFAST extends the widely used 21cmFAST code by incorporating state-of-the-art models of galaxy formation and evolution. It accurately describes the metagalactic radiation background and 21 cm line signal tracing the neutral intergalactic medium through self-consistent photoionization modeling and stellar population synthesis. We demonstrate the code's validity by comparing its predictions with observed evolution of cosmic star formation, IGM neutral fraction, and metal enrichment. LIMFAST also simulates LIM signals of various lines and allows for testing of different modeling aspects when applied to observational data.
ASTROPHYSICAL JOURNAL
(2023)
Article
Biochemical Research Methods
Kaixuan Xiao, Yu Wang, Kangning Dong, Shihua Zhang
Summary: Imaging mass spectrometry (IMS) is a powerful tool for spatial metabolomics, but the complexity of IMS data hinders biomarker acquisition and specific activity studies. To address this, we introduce SmartGate, an AI tool that enables automatic peak selection and spatial structure identification. SmartGate significantly improves spatial segmentation accuracy and biomarker identification based on tissue structure-guided differential analysis.
BRIEFINGS IN BIOINFORMATICS
(2023)
Article
Mechanics
Cengizhan Durucan, Humeyra Sahin, Ayse Rusen Durucan
Summary: This study aims to develop a new seismic intensity measure based on the shape of the response spectrum and the spectral acceleration at the fundamental elastic period of the structure, specifically for short period reinforced concrete structures subjected to near-fault pulse-like ground motion records. The proposed measure shows improved performance compared to other evaluated measures.
MECHANICS BASED DESIGN OF STRUCTURES AND MACHINES
(2023)
Article
Automation & Control Systems
Kun Li, Yingchao Zhang, Yuanlu Li
Summary: Peak detection is crucial in chemical spectrum data processing, but current methods are prone to false positives, leading to erroneous scientific findings. Manual detection is time-consuming, so this paper proposes a novel approach using a multi-scale convolution bi-directional LSTM attention depth network. Experimental results demonstrate an accuracy rate of 89.67% in real LC-MS data, making it a powerful tool for automatic identification of false peaks.
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS
(2023)
Article
Biochemical Research Methods
Pieter Verschaffelt, James Collier, Alexander Botzki, Lennart Martens, Peter Dawyndt, Bart Mesuere
Summary: The Unipept Visualizations library is a JavaScript package that generates interactive visualizations of both hierarchical and non-hierarchical quantitative data, with support for different visualizations and utilizing the D3.js library.
Article
Biochemical Research Methods
Kay Schallert, Pieter Verschaffelt, Bart Mesuere, Dirk Benndorf, Lennart Martens, Tim Van den Bossche
Summary: In metaproteomics, the protein inference problem is more challenging than in single-species proteomics. To address this issue, we developed a tool called Pout2Prot, which converts Percolator output files into protein group output files that can be used with Prophane.
JOURNAL OF PROTEOME RESEARCH
(2022)
Article
Biochemistry & Molecular Biology
Mihaly Varadi, Stephen Anyango, David Armstrong, John Berrisford, Preeti Choudhary, Mandar Deshpande, Nurul Nadzirin, Sreenath S. Nair, Lukas Pravda, Ahsan Tanweer, Bissan Al-Lazikani, Claudia Andreini, Geoffrey J. Barton, David Bednar, Karel Berka, Tom Blundell, Kelly P. Brock, Jose Maria Carazo, Jiri Damborsky, Alessia David, Sucharita Dey, Roland Dunbrack, Juan Fernandez Recio, Franca Fraternali, Toby Gibson, Manuela Helmer-Citterich, David Hoksza, Thomas Hopf, David Jakubec, Natarajan Kannan, Radoslav Krivak, Manjeet Kumar, Emmanuel D. Levy, Nir London, Jose Ramon Macias, Madhusudhan M. Srivatsan, Debora S. Marks, Lennart Martens, Stuart A. McGowan, Jake E. McGreig, Vivek Modi, R. Gonzalo Parra, Gerardo Pepe, Damiano Piovesan, Jaime Prilusky, Valeria Putignano, Leandro G. Radusky, Pathmanaban Ramasamy, Atilio O. Rausch, Nathalie Reuter, Luis A. Rodriguez, Nathan J. Rollins, Antonio Rosato, Luis Serrano, Gulzar Singh, Petr Skoda, Carlos Oscar S. Sorzano, Jan Stourac, Joanna Sulkowska, Radka Svobodova, Natalia Tichshenko, Silvio C. E. Tosatto, Wim Vranken, Mark N. Wass, Dandan Xue, Daniel Zaidman, Janet Thornton, Michael Sternberg, Christine Orengo, Sameer Velankar
Summary: PDBe-KB is an open collaboration platform that aims to integrate functional and biophysical annotations from world-leading specialist data resources, serving the Protein Data Bank. By developing standardized data exchange formats and integrating functional annotations from partner resources, PDBe-KB aims to place macromolecular structure data in a biological context and provide valuable biological insights.
NUCLEIC ACIDS RESEARCH
(2022)
Article
Biochemical Research Methods
Genet Abay Shiferaw, Ralf Gabriels, Robbin Bouwmeester, Tim Van den Bossche, Elien Vandermarliere, Lennart Martens, Pieter-Jan Volders
Summary: Maintaining high sensitivity while limiting false positives in peptide identification from mass spectrometry data is a key challenge. This study investigates the effects of integrating the machine learning-based postprocessor Percolator into the spectral library searching tool COSS.
JOURNAL OF PROTEOME RESEARCH
(2022)
Article
Biochemical Research Methods
Arthur Declercq, Robbin Bouwmeester, Aurelie Hirschler, Christine Carapito, Sven Degroeve, Lennart Martens, Ralf Gabriels
Summary: Immunopeptidomics aims to identify MHC-presented peptides on cells for anti-cancer vaccine development. However, existing data analysis pipelines have difficulty identifying nontryptic peptides. In this study, a retrained (MSPIP)-P-2 model improved predictions for both immuno-peptides and tryptic peptides. The integration of the new models with DeepLC and Percolator in MS(2)Rescore increased spectrum identification rate and unique identified peptides compared to standard Percolator rescoring, and outperformed current immunopeptide-specific identification approaches.
MOLECULAR & CELLULAR PROTEOMICS
(2022)
Article
Biochemical Research Methods
Elke Debrie, Milan Malfait, Ralf Gabriels, Arthur Declerq, Adriaan Sticker, Lennart Martens, Lieven Clement
Summary: Reliable peptide identification is crucial in MS-based proteomics, and the TDA method is widely used for estimating the FDR. However, the assumptions of TDA are often not verified in practice, which can result in poor FDR control and negatively impact downstream data analysis. To address this issue, the TargetDecoy package is introduced, providing necessary functionality to assess the quality and assumptions of TDA for a given set of PSMs.
JOURNAL OF PROTEOME RESEARCH
(2023)
Article
Biochemical Research Methods
Ralf Gabriels, Arthur Declercq, Robbin Bouwmeester, Sven Degroeve, Lennart Martens
Summary: There are various output file formats from proteomics search engines, but the lack of standardized formats makes it difficult to process peptide-spectrum matches (PSMs) and PSM files downstream. This article presents psm_utils, a Python package that can handle various PSM file formats and provides a unified and user-friendly interface. It includes a Python API, a command line interface, and a web application for interconverting PSM files and retrieving basic PSM statistics.
JOURNAL OF PROTEOME RESEARCH
(2023)
Article
Biochemical Research Methods
Benjamin A. Neely, Viktoria Dorfer, Lennart Martens, Isabell Bludau, Robbin Bouwmeester, Sven Degroeve, Eric W. Deutsch, Siegfried Gessulat, Lukas Kaell, Pawel Palczynski, Samuel H. Payne, Tobias Greisager Rehfeldt, Tobias Schmidt, Veit Schwaemmle, Julian Uszkoreit, Juan Antonio Vizcaino, Mathias Wilhelm, Magnus Palmblad
Summary: In recent years, machine learning has made significant progress in modeling mass spectrometry data for proteomics analysis. A workshop was conducted to evaluate and explore machine learning applications in multidimensional mass spectrometry-based proteomics analysis. The workshop helped identify knowledge gaps, define needs, and discuss the possibilities, challenges, and future opportunities. The summary of the discussions conveys excitement about the potential of machine learning in proteomics and aims to inspire future research.
JOURNAL OF PROTEOME RESEARCH
(2023)
Article
Biochemical Research Methods
Tine Claeys, Maxime Menu, Robbin Bouwmeester, Kris Gevaert, Lennart Martens
Summary: Using data from 183 public human data sets, a machine learning model was trained to identify tissue and cell-type specific protein patterns. The model achieved high accuracy in predicting tissues (98%) and cell types (99%) based on protein abundance. The results provide valuable insights into tissue-specific proteins and can be applied to various downstream applications.
JOURNAL OF PROTEOME RESEARCH
(2023)
Article
Biochemical Research Methods
Pieter Verschaffelt, Alessandro Tanca, Marcello Abbondio, Tim van den Bossche, Tibo Vande Moortele, Peter Dawyndt, Lennart Martens, Bart Mesuere
Summary: Unipept Desktop 2.0 is the latest version of the Unipept Desktop tool, which now supports the analysis of metaproteogenomics datasets. It allows for the automatic construction of targeted protein reference databases for improved taxonomic and functional resolution. By limiting the proteins in the database, (meta)proteogenomic analyses can also be performed with better control and privacy. A case study using human gut metaproteome dataset and matched 16S rRNA gene sequencing data is presented as a proof of concept.
JOURNAL OF PROTEOME RESEARCH
(2023)
Article
Multidisciplinary Sciences
Tine Claeys, Tim van den Bossche, Yasset Perez-Riverol, Kris Gevaert, Juan Antonio Vizcaino, Lennart Martens
Summary: Public proteomics data often lack essential metadata, but lesSDRF provides a tool to simplify the process of metadata annotation and ensure that the data has lasting impact beyond its initial publication.
NATURE COMMUNICATIONS
(2023)
Article
Biochemical Research Methods
Ralf Gabriels, Arthur Declercq, Robbin Bouwmeester, Sven Degroeve, Lennart Martens
Summary: This study introduces a Python package called psm_utils, which can read and write various proteomics search engine output file formats and handle peptide-spectrum matches (PSMs) and PSM lists in a unified data structure. The package includes a Python API and command line interface, as well as a user-friendly web application for converting PSM files and retrieving basic PSM statistics.
JOURNAL OF PROTEOME RESEARCH
(2023)