Article
Immunology
Yue Shen, Jerry M. Parks, Jeremy C. Smith
Summary: This study presents a method for classifying HLA class I molecules using three-dimensional structures, which improves the breadth, accuracy, stability, and flexibility of classification. The findings show that structural similarity is highly correlated with peptide binding specificity. This new classification is useful in peptide-based vaccine development and HLA-disease association studies.
JOURNAL OF IMMUNOLOGY
(2023)
Article
Immunology
Paula Ruibal, Kees L. M. C. Franken, Krista E. vzn Meijgaarden, Marjolein van Wolfswinkel, Ian Derksen, Ferenc A. Scheeren, George M. C. Janssen, Peter A. van Veelen, Charlotte Sarfas, Andrew D. White, Sally A. Sharpe, Fabrizio Palmieri, Linda Petrone, Delia Goletti, Thomas Abeel, Tom H. M. Ottenhoff, Simone A. Joosten
Summary: Tuberculosis remains a deadly infectious disease worldwide, causing significant social and economic burden. This study developed an improved algorithm to identify Mycobacterium tuberculosis-derived peptides that can activate specific HLA-E-restricted T cells.
JOURNAL OF IMMUNOLOGY
(2022)
Article
Chemistry, Medicinal
Marko Jukic, Sebastjan Kralj, Anja Kolaric, Urban Bren
Summary: Peptides, as active ingredients of drugs and important tools in medical research, pose a challenge for computational methods. We propose an in silico workflow using CmDock to generate and prioritize peptide libraries, and successfully identified tetrapeptide ligands that bind to antibody Fc regions. Our results align with existing scientific literature and we suggest a developing in silico library design workflow to overcome the combinatorial problem of in vitro peptide libraries.
Article
Multidisciplinary Sciences
Amir Motmaen, Justas Dauparas, Minkyung Baek, Mohamad H. Abedi, David Baker, Philip Bradley
Summary: This study develops a model for predicting peptide-binding proteins and peptide-MHC interactions by adding a classifier on top of the AlphaFold network. The model shows strong generalization and excellent performance.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2023)
Review
Biochemical Research Methods
Meng Wang, Lukasz Kurgan, Min Li
Summary: HLA-I molecules bind intracellular peptides and present them to T cells for immune recognition. Predicting peptides that bind HLA-I molecules is crucial for immunotherapy. We survey 27 tools, evaluating their input/output characteristics, predictive models, and performance. Results show accuracy and meaningful binding motifs in all methods, but no universal superiority. Our comprehensive analysis provides information to identify suitable tools and design future methods.
BRIEFINGS IN BIOINFORMATICS
(2023)
Article
Immunology
Sarah Hall-Swan, Jared Slone, Mauricio M. Rigo, Dinler A. Antunes, Gregory Lizee, Lydia E. Kavraki
Summary: PepSim is a method for predicting T-cell cross-reactivity based on the structural and biochemical similarity of pHLAs. It accurately separates cross-reactive from non-crossreactive pHLAs in diverse datasets, making it a valuable tool for designing safe and effective T-cell immunotherapies.
FRONTIERS IN IMMUNOLOGY
(2023)
Article
Biochemical Research Methods
Qitong Yuan, Keyi Chen, Yimin Yu, Nguyen Quoc Khanh Le, Matthew Chin Heng Chua
Summary: This study proposes a computational method to predict anticancer peptides (ACPs) using sequence information. It utilizes deep learning and machine learning algorithms, including bidirectional long short-term memory (BiLSTM), convolutional neural network (CNN), and Light Gradient Boosting Machine (LightGBM), to classify ACPs. The final model achieves an accuracy of 0.7895, sensitivity of 0.8153, and specificity of 0.7676, surpassing state-of-the-art studies by at least 2% in all metrics. This paper presents a novel and potentially more effective method for ACP prediction.
BRIEFINGS IN BIOINFORMATICS
(2023)
Article
Immunology
Martin Strazar, Jihye Park, Jennifer G. Abelin, Hannah B. Taylor, Thomas K. Pedersen, Damian R. Plichta, Eric M. Brown, Basak Eraslan, Yuan-Mao Hung, Kayla Ortiz, Karl R. Clauser, Steven A. Carr, Ramnik J. Xavier, Daniel B. Graham
Summary: Researchers used monoallelic immunopeptidomics to identify 358,024 HLA-II binders and developed the deep learning model CAPTAn to predict peptide antigens based on their affinity to HLA-II and full sequence of their source proteins. CAPTAn was instrumental in discovering prevalent T cell epitopes from bacteria in the human microbiome and a pan-variant epitope from SARS-CoV-2. CAPTAn and associated datasets provide a resource for antigen discovery and unraveling genetic associations of HLA alleles with immunopathologies.
Article
Chemistry, Medicinal
Shivani Bansal, Kiem Vu, Ruiwu Liu, Yousif Ajena, Wenwu Xiao, Suvidha M. Menon, Amelia Bennett, Angie Gelli, Kit S. Lam
Summary: This work describes the discovery of bead-bound fungal giant unilamellar vesicles (GUVs) over mammalian GUVs, and the optimization of a peptide, K-oLBF127, with higher antifungal activity, lower hemolytic activity, and cytotoxicity. Animal experiments showed that K-oLBF127 has the potential to effectively reduce fungal burden in vivo.
ACS INFECTIOUS DISEASES
(2022)
Article
Immunology
Lihua Deng, Cedric Ly, Sina Abdollahi, Yu Zhao, Immo Prinz, Stefan Bonn
Summary: This study provides a general method for collecting and preprocessing TCR-pMHC binding data and generates comprehensive datasets for comparing prediction models. The performance evaluation of five deep learning models reveals that they have limitations in generalization and robustness. These findings suggest that TCR-pMHC binding prediction remains challenging and requires further improvement in data quality and algorithmic approaches.
FRONTIERS IN IMMUNOLOGY
(2023)
Article
Oncology
Xin Liu, Yixiang Xu, Wei Xiong, Bingnan Yin, Yuqian Huang, Junjun Chu, Changsheng Xing, Chen Qian, Yang Du, Tianhao Duan, Helen Y. Wang, Ningyan Zhang, John S. Yu, Zhiqiang An, Rongfu Wang
Summary: This study developed a TCR-like antibody and CAR-T cells to recognize intracellular proteins as a target for cancer immunotherapy. The specificity and antitumor activity of the antibody and CAR-T cells were demonstrated both in vitro and in vivo, providing a novel approach for the development of cancer immunotherapeutics.
JOURNAL FOR IMMUNOTHERAPY OF CANCER
(2022)
Article
Biochemical Research Methods
Yilin Ye, Jian Wang, Yunwan Xu, Yi Wang, Youdong Pan, Qi Song, Xing Liu, Ji Wan
Summary: Our study presents a pan-allele HLA-peptide binding prediction framework-MATHLA, which integrates bi-directional long short-term memory network and multiple head attention mechanism. This model shows better prediction accuracy in both fivefold cross-validation test and independent test dataset. Additionally, the model outperforms existing tools in predicting longer ligands ranging from 11 to 15 amino acids and demonstrates significant improvement in HLA-C-peptide-binding prediction.
BMC BIOINFORMATICS
(2021)
Article
Biochemical Research Methods
Yaqi Zhang, Gancheng Zhu, Kewei Li, Fei Li, Lan Huang, Meiyu Duan, Fengfeng Zhou
Summary: This study presents a novel HLAB feature engineering algorithm for detecting HLA-I binding peptides using natural language processing and deep neural networks. The experimental results show that the proposed algorithm outperforms existing methods in predicting peptides binding to specific HLA alleles, achieving the best performance in most prediction tasks.
BRIEFINGS IN BIOINFORMATICS
(2022)
Article
Biochemical Research Methods
Shutao Mei, Fuyi Li, Dongxu Xiang, Rochelle Ayala, Pouya Faridi, Geoffrey Webb, Patricia T. Illing, Jamie Rossjohn, Tatsuya Akutsu, Nathan P. Croft, Anthony W. Purcell, Jiangning Song
Summary: Neopeptide-based immunotherapy is a promising approach for cancer treatment, relying on the recognition of neopeptides by CD8(+) T cells through binding to HLA-I molecules. Anthem, an accurate HLA-I binding prediction tool, provides a user-friendly framework for developing customizable prediction models. Extensive evaluations show that Anthem achieves similar or higher AUC values compared to other contemporary tools, offering non-expert users a unique opportunity to analyze their own datasets.
BRIEFINGS IN BIOINFORMATICS
(2021)
Article
Multidisciplinary Sciences
Leidy Pedraza, Milena Camargo, Darwin A. Moreno-Perez, Ricardo Sanchez, Luisa Del Rio-Ospina, Indira M. Baez-Murcia, Manuel E. Patarroyo, Manuel A. Patarroyo
Summary: This study identified 16 HLA-DRB1 alleles in Colombian women that significantly influence the outcome of Chlamydia trachomatis infection. DRB1*08:02:01G and DRB1*12:01:01G were related to infection-promoting events, while only the DQB1*05:03:01G allele related to clearance/persistence events. Women homozygous for HLA-DRB1 alleles were associated with lower clearance probability and/or early occurrence of persistence.
SCIENTIFIC REPORTS
(2021)
Article
Immunology
Saghar Kaabinejadian, Carolina Barra, Bruno Alvarez, Hooman Yari, William H. Hildebrand, Morten Nielsen
Summary: Mass spectrometry-based immunopeptidomics is an important technique in biomedical applications, but the complexity of the data and the lack of appropriate tools have hindered its large-scale application. In this study, a new tool called MHCMotifDecon is presented, which accurately deconvolutes immunopeptidome datasets and helps identify and characterize HLA binding motifs, thus facilitating the discovery of new T cell targets.
FRONTIERS IN IMMUNOLOGY
(2022)
Article
Multidisciplinary Sciences
Zeynep Kosaloglu-Yalcin, Jenny Lee, Jason Greenbaum, Stephen P. Schoenberger, Aaron Miller, Young J. Kim, Alessandro Sette, Morten Nielsen, Bjoern Peters
Summary: The prediction accuracy of antigen epitopes can be further improved by considering the abundance levels of peptides' source proteins. By incorporating biophysical principles, existing MHC binding prediction tools, and abundance estimates of source proteins, a function was derived to estimate the likelihood of a peptide to be an MHC class I ligand. The use of proteomic data showed the highest performance in improving epitope predictions.
Review
Biochemical Research Methods
Morten Nielsen, Nicola Ternette, Carolina Barra
Summary: This article discusses the concept, applications, and challenges of immunopeptidome and emphasizes the benefits and limitations of liquid chromatography-tandem mass spectrometry (MS) in obtaining large-scale immunopeptidome data sets. It highlights the importance of refined and highly optimized machine learning approaches for accurate analysis and interpretation of the data. Furthermore, it showcases the use of MS-immunopeptidomics data in improving the accuracy of immunoinformatics prediction methods and demonstrates the synergistic combination of MS experiments and in silico models for optimal antigen discovery.
EXPERT REVIEW OF PROTEOMICS
(2022)
Article
Biochemistry & Molecular Biology
Magnus Haraldson Hoie, Erik Nicolas Kiehl, Bent Petersen, Morten Nielsen, Ole Winther, Henrik Nielsen, Jeppe Hallgren, Paolo Marcatili
Summary: Recent advances in machine learning and natural language processing have enabled accurate prediction of protein structures and functions, with NetSurfP-3.0 standing out as a tool with drastically improved runtime and reliable prediction performance.
NUCLEIC ACIDS RESEARCH
(2022)
Article
Multidisciplinary Sciences
Jeppe Sejero Holm, Samuel A. Funt, Annie Borch, Kamilla Kjaergaard Munk, Anne-Mette Bjerregaard, James L. Reading, Colleen Maher, Ashley Regazzi, Phillip Wong, Hikmat Al-Ahmadie, Gopa Iyer, Tripti Tamhane, Amalie Kai Bentzen, Nana Overgaard Herschend, Susan De Wolf, Alexandra Snyder, Taha Merghoub, Jedd D. Wolchok, Morten Nielsen, Jonathan E. Rosenberg, Dean F. Bajorin, Sine Reker Hadrup
Summary: This study demonstrates that the expansion of neoantigen-specific CD8(+) T cells can distinguish between patients with controlled disease and progressive disease in metastatic urothelial carcinoma treated with PD-L1 blockade. Furthermore, the peripheral NARTs derived from patients with disease control exhibit specific cell phenotypes and increased CD39 levels, suggesting their association with treatment response.
NATURE COMMUNICATIONS
(2022)
Article
Biochemistry & Molecular Biology
Zeynep Kosaloglu-Yalcin, Nina Blazeska, Randi Vita, Hannah Carter, Morten Nielsen, Stephen Schoenberger, Alessandro Sette, Bjoern Peters
Summary: CEDAR is a freely accessible database and analysis resource for cancer epitopes, which are molecular targets recognized by anti-cancer immune cells. Detailed knowledge of cancer epitopes is crucial for understanding and planning cancer prevention, treatment, and immune responses.
NUCLEIC ACIDS RESEARCH
(2023)
Article
Oncology
Loulieta Nazerai, Shona Caroline Willis, Patricio Yankilevich, Luca Di Leo, Francesca Maria Bosisio, Alex Frias, Corine Bertolotto, Jacob Nersting, Maria Thastrup, Soren Buus, Allan Randrup Thomsen, Morten Nielsen, Kristoffer Staal Rohrberg, Kjeld Schmiegelow, Daniela De Zio
Summary: This study used the drug 6TG to induce mutations in tumor cells and increase the level of neoepitopes, enhancing the immune response. 6TG exposure increased tumor mutational burden and reshaped the tumor microenvironment, making the tumors more responsive to immune-checkpoint blockade.
Article
Immunology
Timothy Connelley, Annalisa Nicastri, Tara Sheldrake, Christina Vrettou, Andressa Fisch, Birkir Reynisson, Soren Buus, Adrian Hill, Ivan Morrison, Morten Nielsen, Nicola Ternette
Summary: This study used immunopeptidomics to study the repertoire of peptides presented by Theileria parva-infected cells. A comprehensive dataset of 74 BoLA-I and 15 BoLA-DR-presented peptides, based on different BoLA genotypes, was generated. The study demonstrated the utility of immunopeptidomics as a method to identify novel T-cell antigens for T. parva.
Article
Biochemistry & Molecular Biology
Arnor Sigurdsson, Ioannis Louloudis, Karina Banasik, David Westergaard, Ole Winther, Ole Lund, Sisse Rye Ostrowski, Christian Erikstrup, Ole Birger Vesterager Pedersen, Mette Nyegaard, Soren Brunak, Bjarni J. Vilhjalmsson, Simon Rasmussen
Summary: We developed a deep learning framework for polygenic risk score (PRS) prediction that can handle large-scale genomics data, support multi-task learning, and automatically integrate clinical and biochemical data. The framework demonstrated competitive performance and improved predictions for complex genetic relationships and non-additive genetic effects and epistasis. The model also outperformed traditional linear PRS methods for Type 1 Diabetes.
NUCLEIC ACIDS RESEARCH
(2023)
Article
Biology
Helle Rus Povlsen, Amalie Kai Bentzen, Mohammad Kadivar, Leon Eyrich Jessen, Sine Reker Hadrup, Morten Nielsen, K. Christopher Garcia
Summary: Novel single-cell-based technologies enable high-throughput matching of T cell receptor (TCR) sequences with their cognate peptide-MHC recognition motif. A data-driven method called ITRAP is proposed to filter out likely artifacts and generate large sets of TCR-pMHC sequence data with high specificity and sensitivity. This approach has been validated in virus-specific T cell responses across multiple healthy donors.
Article
Biology
Jonas Birkelund Nilsson, Saghar Kaabinejadian, Hooman Yari, Bjoern Peters, Carolina Barra, Loren Gragert, William Hildebrand, Morten Nielsen
Summary: This study addresses the prediction of HLA-DQ antigen presentation and the contribution of trans-only variants in shaping the HLA-DQ immunopeptidome. By integrating immunoinformatics data mining models with mass spectrometry immunopeptidomics data, the study demonstrates improved predictive power and molecular coverage for models trained with novel HLA-DQ data. The study also reveals the limited contribution of trans-only HLA-DQ variants to the overall HLA-DQ immunopeptidome.
COMMUNICATIONS BIOLOGY
(2023)
Article
Immunology
Sandra Eltschkner, Samantha Mellinger, Soren Buus, Morten Nielsen, Kajsa M. M. Paulsson, Karin Lindkvist-Petersson, Helena Westerdahl
Summary: Long-distance migratory animals such as birds and bats have evolved a unique adaptive immunity with highly duplicated Major Histocompatibility Complex (MHC) genes to withstand diverse pathogens. A study on the MHC class I protein, Acar3, from the great reed warbler reveals a peculiar peptide-binding mode that potentially facilitates interactions with innate immune receptors. The investigation highlights the importance of studying the immune system of wild animals to uncover unique immune mechanisms absent in humans and model organisms.
FRONTIERS IN IMMUNOLOGY
(2023)
Article
Mathematical & Computational Biology
Yuchen Li, Peter Wad Sackett, Morten Nielsen, Carolina Barra
Summary: This article introduces a new protein allergenicity prediction method, which introduces MHC presentation propensity as a novel feature to overcome the limitations of previous methods in accurately predicting allergenicity when similarity diminishes.
BIOINFORMATICS ADVANCES
(2023)
Meeting Abstract
Physiology
M. S. Khilji, P. Faridi, E. Pinheiro-Machado, C. Hoefner, T. Dahlby, R. Aranha, S. Buus, M. Nielsen, J. Klusek, T. Mandrup-Poulsen, K. Pandey, A. W. Purcell, M. T. Marzec
Article
Medicine, Research & Experimental
Nikolaj Pagh Kristensen, Christina Heeke, Siri A. Tvingsholm, Annie Borch, Arianna Draghi, Michael D. Crowther, Ibel Carri, Kamilla K. Munk, Jeppe Sejero Holm, Anne-Mette Bjerregaard, Amalie Kai Bentzen, Andrea M. Marquard, Zoltan Szallasi, Nicholas McGranahan, Rikke Andersen, Morten Nielsen, Goran B. Jonsson, Marco Donia, Inge Marie Svane, Sine Reker Hadrup
Summary: This study examined the frequency of neoepitope-specific CD8(+) T cells in tumor-infiltrating lymphocyte (TIL) infusion products and blood samples from melanoma patients who received adoptive cell therapy (ACT). The results showed that the frequency of neoepitope-specific CD8(+) T cells correlated with patient survival, and these cells were unique to responders of TIL-ACT. Additionally, a transcriptional signature for lymphocyte activity within the tumor microenvironment was associated with a higher frequency of neoepitope-specific CD8(+) T cells in the infusion product.
JOURNAL OF CLINICAL INVESTIGATION
(2022)