ATM-TCR: TCR-Epitope Binding Affinity Prediction Using a Multi-Head Self-Attention Model.
Ontology highlight
ABSTRACT:
SUBMITTER: Cai M
PROVIDER: S-EPMC9299376 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Subcellular localization of messenger RNAs (mRNAs), as a prevalent mechanism, gives precise and efficient control for the translation process. There is mounting evidence for the important roles of this process in a variety of cellular events. Computational methods for mRNA subcellular localization prediction provide a useful approach for studying mRNA functions. However, few computational methods were designed for mRNA subcellular localization prediction and their performance have room for improvement. Especially, there is still no available tool to predict for mRNAs that have multiple localization annotations. In this paper, we propose a multi-head self-attention method, DM3Loc, for multi-label mRNA subcellular localization prediction. Evaluation results show that DM3Loc outperforms existing methods and tools in general. Furthermore, DM3Loc has the interpretation ability to analyze RNA-binding protein motifs and key signals on mRNAs for subcellular localization. Our analyses found hundreds of instances of mRNA isoform-specific subcellular localizations and many significantly enriched gene functions for mRNAs in different subcellular localizations.
| S-EPMC8096227 | biostudies-literature
Project description:Drug-target interaction (DTI) prediction has drawn increasing interest due to its substantial position in the drug discovery process. Many studies have introduced computational models to treat DTI prediction as a regression task, which directly predict the binding affinity of drug-target pairs. However, existing studies (i) ignore the essential correlations between atoms when encoding drug compounds and (ii) model the interaction of drug-target pairs simply by concatenation. Based on those observations, in this study, we propose an end-to-end model with multiple attention blocks to predict the binding affinity scores of drug-target pairs. Our proposed model offers the abilities to (i) encode the correlations between atoms by a relation-aware self-attention block and (ii) model the interaction of drug representations and target representations by the multi-head attention block. Experimental results of DTI prediction on two benchmark datasets show our approach outperforms existing methods, which are benefit from the correlation information encoded by the relation-aware self-attention block and the interaction information extracted by the multi-head attention block. Moreover, we conduct the experiments on the effects of max relative position length and find out the best max relative position length value $k \in \{3, 5\}$. Furthermore, we apply our model to predict the binding affinity of Corona Virus Disease 2019 (COVID-19)-related genome sequences and $3137$ FDA-approved drugs.
| S-EPMC8083346 | biostudies-literature
Project description:This SuperSeries is composed of the SubSeries listed below.
2022-04-26 | GSE201485 | GEO
Project description:The dual functional lncRNAs have been intensively studied and identified to be involved in various fundamental cellular processes recently. It is essential to understand in which context when a dual functional lncRNA serves as a non-coding RNA or a template for coding peptide, particularly in some pathological conditions. However, apart from time consuming and cell type specific experiments, there is virtually no in-silico method for predicting the identity of dual functional lncRNAs. Here, we developed a deep-learning model with multi-head self-attention mechanism, LncReader, to identify dual functional lncRNAs based on their sequence, physicochemical and secondary structural features. Our data demonstrated that LncReader showed multiple advantage compared to various classical machine learning methods. Moreover, to obtain independent in-house datasets for robust testing, mass spectrometry proteomics combined with RNA-seq were applied in four leukemia cell lines. Remarkably, LncReader achieved the best performance among all these datasets. Therefore, LncReader provides a sophisticated and practical tool that enables fast dual functional lncRNAs identification.
2023-04-26 | PXD030066 | Pride
Project description:Viruses like HIV and SIV escape from containment by CD8(+) T lymphocytes through generating mutations that interfere with epitope peptide:MHC class I binding. However, mutations in some viral epitopes are selected for that have no impact on this binding. We explored the mechanism underlying the evolution of such epitopes by studying CD8(+) T lymphocyte recognition of a dominant Nef epitope of SIVmac251 in infected Mamu-A*02(+) rhesus monkeys. Clonal analysis of the p199RY-specific CD8(+) T lymphocyte repertoire in these monkeys indicated that identical T cell clones were capable of recognizing wild-type (WT) and mutant epitope sequences. However, we found that the functional avidity of these CD8(+) T lymphocytes for the mutant peptide:Mamu-A*02 complex was diminished. Using surface plasmon resonance to measure the binding affinity of the p199RY-specific TCR repertoire for WT and mutant p199RY peptide:Mamu-A*02 monomeric complexes, we found that the mutant p199RY peptide:Mamu-A*02 complexes had a lower affinity for TCRs purified from CD8(+) T lymphocytes than did the WT p199RY peptide:Mamu-A*02 complexes. These studies demonstrated that differences in TCR affinity for peptide:MHC class I ligands can alter functional p199RY-specific CD8(+) T lymphocyte responses to mutated epitopes, decreasing the capacity of these cells to contain SIVmac251 replication.
| S-EPMC3175636 | biostudies-other
Project description:The dual functional lncRNAs have been intensively studied and identified to be involved in various fundamental cellular processes recently. It is essential to understand in which context when a dual functional lncRNA serves as a non-coding RNA or a template for coding peptide, particularly in some pathological conditions. However, apart from time consuming and cell type specific experiments, there is virtually no in-silico method for predicting the identity of dual functional lncRNAs. Here, we developed a deep-learning model with multi-head self-attention mechanism, LncReader, to identify dual functional lncRNAs based on their sequence, physicochemical and secondary structural features. Our data demonstrated that LncReader showed multiple advantage compared to various classical machine learning methods. Moreover, to obtain independent in-house datasets for robust testing, mass spectrometry proteomics combined with RNA-seq were applied in four leukemia cell lines. Remarkably, LncReader achieved the best performance among all these datasets. Therefore, LncReader provides a sophisticated and practical tool that enables fast dual functional lncRNAs identification.
2021-11-25 | GSE189394 | GEO
Project description:The dual functional lncRNAs have been intensively studied and identified to be involved in various fundamental cellular processes recently. It is essential to understand in which context when a dual functional lncRNA serves as a non-coding RNA or a template for coding peptide, particularly in some pathological conditions. However, apart from time consuming and cell type specific experiments, there is virtually no in-silico method for predicting the identity of dual functional lncRNAs. Here, we developed a deep-learning model with multi-head self-attention mechanism, LncReader, to identify dual functional lncRNAs based on their sequence, physicochemical and secondary structural features. Our data demonstrated that LncReader showed multiple advantage compared to various classical machine learning methods. Moreover, to obtain independent in-house datasets for robust testing, mass spectrometry proteomics combined with RNA-seq were applied in four leukemia cell lines. Remarkably, LncReader achieved the best performance among all these datasets. Therefore, LncReader provides a sophisticated and practical tool that enables fast dual functional lncRNAs identification.
2022-04-26 | GSE201484 | GEO
Project description:Background Accurate prediction of protein–ligand binding affinity is important for lowering the overall cost of drug discovery in structure-based drug design. For accurate predictions, many classical scoring functions and machine learning-based methods have been developed. However, these techniques tend to have limitations, mainly resulting from a lack of sufficient energy terms to describe the complex interactions between proteins and ligands. Recent deep-learning techniques can potentially solve this problem. However, the search for more efficient and appropriate deep-learning architectures and methods to represent protein–ligand complex is ongoing. Results In this study, we proposed a deep-neural network model to improve the prediction accuracy of protein–ligand complex binding affinity. The proposed model has two important features, descriptor embeddings with information on the local structures of a protein–ligand complex and an attention mechanism to highlight important descriptors for binding affinity prediction. The proposed model performed better than existing binding affinity prediction models on most benchmark datasets. Conclusions We confirmed that an attention mechanism can capture the binding sites in a protein–ligand complex to improve prediction performance. Our code is available at https://github.com/Blue1993/BAPA. Supplementary Information The online version contains supplementary material available at 10.1186/s12859-021-04466-0.
| S-EPMC8576937 | biostudies-literature
Project description:BACKGROUND:Affinity-optimized T cell receptor (TCR)-engineered lymphocytes targeting tumor antigens can mediate potent antitumor responses in cancer patients, but also bear substantial risks for off-target toxicities. Most preclinical studies have focused on T cell responses to antigen-specific stimulation. In contrast, little is known on the regulation of T cell responsiveness through continuous TCR triggering and consequent tonic signaling. Here, we addressed the question whether increasing the TCR affinity can lead to chronic interactions occurring directly between TCRs and MHC-(self) molecules, which may modulate the overall functional potency of tumor-redirected CD8 T cells. For this purpose, we developed two complementary human CD8 T cell models (i.e. HLA-A2 knock-in and knock-out) engineered with incremental-affinity TCRs to the HLA-A2/NY-ESO-1 tumor antigen. METHODS:The impact of HLA-A2 recognition, depending on TCR affinity, was assessed at the levels of the TCR/CD3 complex, regulatory receptors, and signaling, under steady-state conditions and in kinetic studies. The quality of CD8 T cell responses was further evaluated by gene expression and multiplex cytokine profiling, as well as real-time quantitative cell killing, combined with co-culture assays. RESULTS:We found that HLA-A2 per se (in absence of cognate peptide) can trigger chronic activation followed by a tolerance-like state of tumor-redirected CD8 T cells with increased-affinity TCRs. HLA-A2pos but not HLA-A2neg T cells displayed an activation phenotype, associated with enhanced upregulation of c-CBL and multiple inhibitory receptors. T cell activation preceded TCR/CD3 downmodulation, impaired TCR signaling and functional hyporesponsiveness. This stepwise activation-to-hyporesponsive state was dependent on TCR affinity and already detectable at the upper end of the physiological affinity range (KD???1??M). Similar findings were made when affinity-increased HLA-A2neg CD8 T cells were chronically exposed to HLA-A2pos-expressing target cells. CONCLUSIONS:Our observations indicate that sustained interactions between affinity-increased TCR and self-MHC can directly adjust the functional potential of T cells, even in the absence of antigen-specific stimulation. The observed tolerance-like state depends on TCR affinity and has therefore potential implications for the design of affinity-improved TCRs for adoptive T cell therapy, as several engineered TCRs currently used in clinical trials share similar affinity properties.
| S-EPMC6833194 | biostudies-literature
Project description:Adoptive cellular therapy (ACT) using T-cell receptor (TCR)-engineered lymphocytes holds promise for eradication of disseminated tumors but also an inherent risk of pathologic autoimmunity if targeted antigens or antigenic mimics are expressed by normal tissues. We evaluated whether modulating TCR affinity could allow CD8+ T cells to control tumor outgrowth without inducing concomitant autoimmunity in a preclinical murine model of ACT. RIP-mOVA mice express a membrane-bound form of chicken ovalbumin (mOVA) as a self-antigen in kidney and pancreas. Such mice were implanted with OVA-expressing ID8 ovarian carcinoma cells and subsequently treated with CD8+ T lymphocytes (CTL) expressing either a high-affinity (OT-I) or low-affinity (OT-3) OVA-specific TCR. The effects on tumor growth versus organ-specific autoimmunity were subsequently monitored. High-affinity OT-I CTLs underwent activation and proliferation in both tumor-draining and pancreatic lymph nodes, leading to both rapid eradication of ID8-OVA tumors and autoimmune diabetes in all treated mice. Remarkably, the low-affinity OT-3 T cells were activated only by tumor-derived antigen and mediated transient regression of ID8-OVA tumors without concomitant autoimmunity. The OT-3 cells eventually upregulated inhibitory receptors PD-1, TIM-3, and LAG-3 and became functionally unresponsive, however, allowing the tumors in treated mice to reestablish progressive growth. Antibody-mediated blockade of the inhibitory receptors prevented exhaustion and allowed tumor clearance, but these mice also developed autoimmune diabetes. The findings reveal that low-affinity TCRs can mediate tumor regression and that functional avidity can discriminate between tumor-derived and endogenous antigen, while highlighting the risks involved in immune-checkpoint blockade on endogenous self-reactive T cells.
| S-EPMC7793606 | biostudies-literature