PSSMHCpan: a novel PSSM-based software for predicting class I peptide-HLA binding affinity

Synthetic multiepitope neoantigen DNA vaccine for personalized cancer immunotherapy

Neoantigen-based personalized vaccination has emerged as a viable method for tumor immunotherapy. Here we set up a DNA-based neoantigen vaccine platform with comprehensive identification of individual somatic mutations using whole-exome sequencing (WES) and RNA-seq, bioinformatic prediction of neo-epitopes, dendritic cell (DC)-based efficacy prevalidation of vaccine candidates, optimization of the DNA vaccine and its nanocarrier and adjuvant, and preparation of a liposome-encapsulated multiepitope DNA vaccine. The DNA vaccine was efficiently uptaken by DCs and induced effective immune response against mouse melanoma cells, leading to significant inhibition of melanoma tumor growth and reduction of lung metastasis in a mouse model. Numerous intratumoral infiltrated CD8+ T-cells with specific in vitro killing ability towards melanoma cells were identified. Our study offers evidence that a multiepitope neoantigen DNA vaccine in a nanocarrier can be exploited for personalized tumor immunotherapy and as a reliable prevalidation approach for rapid enrichment of effective neoantigens.

Predicting MHC-peptide binding affinity by differential boundary tree

Motivation

The prediction of the binding between peptides and major histocompatibility complex (MHC) molecules plays an important role in neoantigen identification. Although a large number of computational methods have been developed to address this problem, they produce high false-positive rates in practical applications, since in most cases, a single residue mutation may largely alter the binding affinity of a peptide binding to MHC which cannot be identified by conventional deep learning methods.

Results

We developed a differential boundary tree-based model, named DBTpred, to address this problem. We demonstrated that DBTpred can accurately predict MHC class I binding affinity compared to the state-of-art deep learning methods. We also presented a parallel training algorithm to accelerate the training and inference process which enables DBTpred to be applied to large datasets. By investigating the statistical properties of differential boundary trees and the prediction paths to test samples, we revealed that DBTpred can provide an intuitive interpretation and possible hints in detecting important residue mutations that can largely influence binding affinity.

Availability and implementation

The DBTpred package is implemented in Python and freely available at: https://github.com/fpy94/DBT.

Supplementary information

Supplementary data are available at Bioinformatics online.

Predicting MHC class I binder: existing approaches and a novel recurrent neural network solution

Major histocompatibility complex (MHC) possesses important research value in the treatment of complex human diseases. A plethora of computational tools has been developed to predict MHC class I binders. Here, we comprehensively reviewed 27 up-to-date MHC I binding prediction tools developed over the last decade, thoroughly evaluating feature representation methods, prediction algorithms and model training strategies on a benchmark dataset from Immune Epitope Database. A common limitation was identified during the review that all existing tools can only handle a fixed peptide sequence length. To overcome this limitation, we developed a bilateral and variable long short-term memory (BVLSTM)-based approach, named BVLSTM-MHC. It is the first variable-length MHC class I binding predictor. In comparison to the 10 mainstream prediction tools on an independent validation dataset, BVLSTM-MHC achieved the best performance in six out of eight evaluated metrics. A web server based on the BVLSTM-MHC model was developed to enable accurate and efficient MHC class I binder prediction in human, mouse, macaque and chimpanzee.

Beyond Tumor Mutation Burden: Tumor Neoantigen Burden as a Biomarker for Immunotherapy and Other Types of Therapy

Immunotherapy has significantly improved the clinical outcome of patients with cancer. However, the immune response rate varies greatly, possibly due to lack of effective biomarkers that can be used to distinguish responders from non-responders. Recently, clinical studies have associated high tumor neoantigen burden (TNB) with improved outcomes in patients treated with immunotherapy. Therefore, TNB has emerged as a biomarker for immunotherapy and other types of therapy. In the present review, the potential application of TNB as a biomarker was evaluated. The methods of neoantigen prediction were summarized and the mechanisms involved in TNB were investigated. The impact of high TNB and increased number of infiltrating immune cells on the efficacy of immunotherapy was also addressed. Finally, the future challenges of TNB were discussed.

Peptide-based therapeutic cancer vaccine: Current trends in clinical application

The peptide‐based therapeutic cancer vaccines have attracted enormous attention in recent years as one of the effective treatments of tumour immunotherapy. Most of peptide‐based vaccines are based on epitope peptides stimulating CD8⁺ T cells or CD4⁺ T helper cells to target tumour‐associated antigens (TAAs) or tumour‐specific antigens (TSAs). Some adjuvants and nanomaterials have been exploited to optimize the efficiency of immune response of the epitope peptide to improve its clinical application. At present, numerous peptide‐based therapeutic cancer vaccines have been developed and achieved significant clinical benefits. Similarly, the combination of peptide‐based vaccines and other therapies has demonstrated a superior efficacy in improving anti‐cancer activity. We delve deeper into the choices of targets, design and screening of epitope peptides, clinical efficacy and adverse events of peptide‐based vaccines, and strategies combination of peptide‐based therapeutic cancer vaccines and other therapies. The review will provide a detailed overview and basis for future clinical application of peptide‐based therapeutic cancer vaccines.

A Proteome-Wide Immunoinformatics Tool to Accelerate T-Cell Epitope Discovery and Vaccine Design in the Context of Emerging Infectious Diseases: An Ethnicity-Oriented Approach

Emerging infectious diseases (EIDs) caused by viruses are increasing in frequency, causing a high disease burden and mortality world-wide. The COVID-19 pandemic caused by the novel SARS-like coronavirus (SARS-CoV-2) underscores the need to innovate and accelerate the development of effective vaccination strategies against EIDs. Human leukocyte antigen (HLA) molecules play a central role in the immune system by determining the peptide repertoire displayed to the T-cell compartment. Genetic polymorphisms of the HLA system thus confer a strong variability in vaccine-induced immune responses and may complicate the selection of vaccine candidates, because the distribution and frequencies of HLA alleles are highly variable among different ethnic groups. Herein, we build on the emerging paradigm of rational epitope-based vaccine design, by describing an immunoinformatics tool (Predivac-3.0) for proteome-wide T-cell epitope discovery that accounts for ethnic-level variations in immune responsiveness. Predivac-3.0 implements both CD8+ and CD4+ T-cell epitope predictions based on HLA allele frequencies retrieved from the Allele Frequency Net Database. The tool was thoroughly assessed, proving comparable performances (AUC ~0.9) against four state-of-the-art pan-specific immunoinformatics methods capable of population-level analysis (NetMHCPan-4.0, Pickpocket, PSSMHCPan and SMM), as well as a strong accuracy on proteome-wide T-cell epitope predictions for HIV-specific immune responses in the Japanese population. The utility of the method was investigated for the COVID-19 pandemic, by performing in silico T-cell epitope mapping of the SARS-CoV-2 spike glycoprotein according to the ethnic context of the countries where the ChAdOx1 vaccine is currently initiating phase III clinical trials. Potentially immunodominant CD8+ and CD4+ T-cell epitopes and population coverages were predicted for each population (the Epitope Discovery mode), along with optimized sets of broadly recognized (promiscuous) T-cell epitopes maximizing coverage in the target populations (the Epitope Optimization mode). Population-specific epitope-rich regions (T-cell epitope clusters) were further predicted in protein antigens based on combined criteria of epitope density and population coverage. Overall, we conclude that Predivac-3.0 holds potential to contribute in the understanding of ethnic-level variations of vaccine-induced immune responsiveness and to guide the development of epitope-based next-generation vaccines against emerging pathogens, whose geographic distributions and populations in need of vaccinations are often well-defined for regional epidemics.

Anthem: a user customised tool for fast and accurate prediction of binding between peptides and HLA class I molecules

Neopeptide-based immunotherapy has been recognised as a promising approach for the treatment of cancers. For neopeptides to be recognised by CD8+ T cells and induce an immune response, their binding to human leukocyte antigen class I (HLA-I) molecules is a necessary first step. Most epitope prediction tools thus rely on the prediction of such binding. With the use of mass spectrometry, the scale of naturally presented HLA ligands that could be used to develop such predictors has been expanded. However, there are rarely efforts that focus on the integration of these experimental data with computational algorithms to efficiently develop up-to-date predictors. Here, we present Anthem for accurate HLA-I binding prediction. In particular, we have developed a user-friendly framework to support the development of customisable HLA-I binding prediction models to meet challenges associated with the rapidly increasing availability of large amounts of immunopeptidomic data. Our extensive evaluation, using both independent and experimental datasets shows that Anthem achieves an overall similar or higher area under curve value compared with other contemporary tools. It is anticipated that Anthem will provide a unique opportunity for the non-expert user to analyse and interpret their own in-house or publicly deposited datasets.

Identification and validation of 174 COVID-19 vaccine candidate epitopes reveals low performance of common epitope prediction tools

The outbreak of SARS-CoV-2 (2019-nCoV) virus has highlighted the need for fast and efficacious vaccine development. Stimulation of a proper immune response that leads to protection is highly dependent on presentation of epitopes to circulating T-cells via the HLA complex. SARS-CoV-2 is a large RNA virus and testing of all of its overlapping peptides in vitro to deconvolute an immune response is not feasible. Therefore HLA-binding prediction tools are often used to narrow down the number of peptides to test. We tested NetMHC suite tools' predictions by using an in vitro peptide-MHC stability assay. We assessed 777 peptides that were predicted to be good binders across 11 MHC alleles in a complex-stability assay and tested a selection of 19 epitope-HLA-binding prediction tools against the assay. In this investigation of potential SARS-CoV-2 epitopes we found that current prediction tools vary in performance when assessing binding stability, and they are highly dependent on the MHC allele in question. Designing a COVID-19 vaccine where only a few epitope targets are included is therefore a very challenging task. Here, we present 174 SARS-CoV-2 epitopes with high prediction binding scores, validated to bind stably to 11 HLA alleles. Our findings may contribute to the design of an efficacious vaccine against COVID-19.

dbPepNeo: a manually curated database for human tumor neoantigen peptides

Neoantigens can function as actual antigens to facilitate tumor rejection, which play a crucial role in cancer immunology and immunotherapy. Emerging evidence revealed that neoantigens can be used to develop personalized, cancer-specific vaccines. To date, large numbers of immunogenomic peptides have been computationally predicted to be potential neoantigens. However, experimental validation remains the gold standard for potential clinical application. Experimentally validated neoantigens are rare and mostly appear scattered among scientific papers and various databases. Here, we constructed dbPepNeo, a specific database for human leukocyte antigen class I (HLA-I) binding neoantigen peptides based on mass spectrometry (MS) validation or immunoassay in human tumors. According to the verification methods of these neoantigens, the collection of peptides was classified as 295 high confidence, 247 medium confidence and 407 794 low confidence neoantigens, respectively. This can serve as a valuable resource to aid further screening for effective neoantigens, optimize a neoantigen prediction pipeline and study T-cell receptor (TCR) recognition. Three applications of dbPepNeo are shown. In summary, this work resulted in a platform to promote the screening and confirmation of potential neoantigens in cancer immunotherapy.

Database URL: www.biostatistics.online/dbPepNeo/.

TruNeo: an integrated pipeline improves personalized true tumor neoantigen identification

Background

Neoantigen-based personal vaccines and adoptive T cell immunotherapy have shown high efficacy as a cancer treatment in clinical trials. Algorithms for the accurate prediction of neoantigens have played a pivotal role in such studies. Some existing bioinformatics methods, such as MHCflurry and NetMHCpan, identify neoantigens mainly through the prediction of peptide-MHC binding affinity. However, the predictive accuracy of immunogenicity of these methods has been shown to be low. Thus, a ranking algorithm to select highly immunogenic neoantigens of patients is needed urgently in research and clinical practice.

Results

We develop TruNeo, an integrated computational pipeline to identify and select highly immunogenic neoantigens based on multiple biological processes. The performance of TruNeo and other algorithms were compared based on data from published literature as well as raw data from a lung cancer patient. Recall rate of immunogenic ones among the top 10-ranked neoantigens were compared based on the published combined data set. Recall rate of TruNeo was 52.63%, which was 2.5 times higher than that predicted by MHCflurry (21.05%), and 2 times higher than NetMHCpan 4 (26.32%). Furthermore, the positive rate of top 10-ranked neoantigens for the lung cancer patient were compared, showing a 50% positive rate identified by TruNeo, which was 2.5 times higher than that predicted by MHCflurry (20%).

Conclusions

TruNeo, which considers multiple biological processes rather than peptide-MHC binding affinity prediction only, provides prioritization of candidate neoantigens with high immunogenicity for neoantigen-targeting personalized immunotherapies.

The co-stimulation of anti-CD28 and IL-2 enhances the sensitivity of ELISPOT assays for detection of neoantigen-specific T cells in PBMC

Neoantigen-based cancer immunotherapies hold the promise of being a truly personalized, effective treatment for diverse cancer types. ELISPOT assays, as a powerful experimental technique, can verify the existence of antigen specific T cells to support basic clinical research and monitor clinical trials. However, despite the high sensitivity of ELISPOT assays, detecting immune responses of neoantigen specific T cells in a patient or healthy donor's PBMCs is still extremely difficult, since the frequency of these T cells can be very low. We developed a novel experimental method, by co-stimulation of T cells with anti-CD28 and IL-2 at the beginning of ELISPOT, to further increase the sensitivity of ELISPOT and mitigate the challenge introduced by low frequency T cells. Under the optimal concentration of 1 μg/ml for anti-CD28 and 1 U/ml for IL-2, an 11.7-fold increase of T cell response against CMV peptide was observed by using our method, and it outperforms other cytokine stimulation alternatives (5-10 folds). We also showed that this method can be effectively applied to detect neoantigen-specific T cells in healthy donors' and a melanoma patient's PBMCs. To the best of our knowledge, this is the first report that the co-stimulation of anti-CD28 and IL-2 is able to significantly improve the sensitivity of ELISPOT assays, indicating that anti-CD28 and IL-2 signaling can act in synergy to lower the T cell activation threshold and trigger more neoantigen-specific T cells.

Recurrent Neoantigens in Colorectal Cancer as Potential Immunotherapy Targets

This study was aimed at investigating the mutations in colorectal cancer (CRC) for recurrent neoantigen identification. A total of 1779 samples with whole exome sequencing (WES) data were obtained from 7 published CRC cohorts. Common HLA genotypes were used to predict the probability of neoantigens at high-frequency mutants in the dataset. Based on the WES data, we not only obtained the most comprehensive CRC mutation landscape so far but also found 1550 mutations which could be identified in at least 5 patients, including KRAS G12D (8%), KRAS G12V (5.8%), PIK3CA E545K (3.5%), PIK3CA H1047R (2.5%), and BMPR2 N583Tfs∗44 (2.8%). These mutations can also be recognized by multiple common HLA molecules in Chinese and TCGA cohort as potential "public" neoantigens. Many of these mutations also have high mutation rates in metastatic pan-cancers, suggesting their value as therapeutic targets in different cancer types. Overall, our analysis provides recurrent neoantigens as potential cancer immunotherapy targets.

Uncovering the Tumor Antigen Landscape: What to Know about the Discovery Process

According to the latest available data, cancer is the second leading cause of death, highlighting the need for novel cancer therapeutic approaches. In this context, immunotherapy is emerging as a reliable first-line treatment for many cancers, particularly metastatic melanoma. Indeed, cancer immunotherapy has attracted great interest following the recent clinical approval of antibodies targeting immune checkpoint molecules, such as PD-1, PD-L1, and CTLA-4, that release the brakes of the immune system, thus reviving a field otherwise poorly explored. Cancer immunotherapy mainly relies on the generation and stimulation of cytotoxic CD8 T lymphocytes (CTLs) within the tumor microenvironment (TME), priming T cells and establishing efficient and durable anti-tumor immunity. Therefore, there is a clear need to define and identify immunogenic T cell epitopes to use in therapeutic cancer vaccines. Naturally presented antigens in the human leucocyte antigen-1 (HLA-I) complex on the tumor surface are the main protagonists in evocating a specific anti-tumor CD8+ T cell response. However, the methodologies for their identification have been a major bottleneck for their reliable characterization. Consequently, the field of antigen discovery has yet to improve. The current review is intended to define what are today known as tumor antigens, with a main focus on CTL antigenic peptides. We also review the techniques developed and employed to date for antigen discovery, exploring both the direct elution of HLA-I peptides and the in silico prediction of epitopes. Finally, the last part of the review analyses the future challenges and direction of the antigen discovery field.

Shared neoantigens: ideal targets for off-the-shelf cancer immunotherapy

Neoantigen, as an important member of tumor-specific antigens, has attracted a great deal of attention as a target for immunotherapy. Neoantigens are potential targets for personalized vaccines and adoptive cell transfer therapies. However, most of the neoantigen-targeted immunotherapies in the process are customized and costly. So, we are inclined to find shared neoantigens suitable for more patients. With the help of existing neoantigen prediction algorithms, we found that the most frequent shared neoantigens occurred in more than 1% of patients for 17 tumor types and the ten most frequent shared neoantigens covered approximately 50% of pancreatic cancer patients, providing a potential list of targets for off-the-shelf immunotherapy.

A comprehensive review and performance evaluation of bioinformatics tools for HLA class I peptide-binding prediction

Human leukocyte antigen class I (HLA-I) molecules are encoded by major histocompatibility complex (MHC) class I loci in humans. The binding and interaction between HLA-I molecules and intracellular peptides derived from a variety of proteolytic mechanisms play a crucial role in subsequent T-cell recognition of target cells and the specificity of the immune response. In this context, tools that predict the likelihood for a peptide to bind to specific HLA class I allotypes are important for selecting the most promising antigenic targets for immunotherapy. In this article, we comprehensively review a variety of currently available tools for predicting the binding of peptides to a selection of HLA-I allomorphs. Specifically, we compare their calculation methods for the prediction score, employed algorithms, evaluation strategies and software functionalities. In addition, we have evaluated the prediction performance of the reviewed tools based on an independent validation data set, containing 21 101 experimentally verified ligands across 19 HLA-I allotypes. The benchmarking results show that MixMHCpred 2.0.1 achieves the best performance for predicting peptides binding to most of the HLA-I allomorphs studied, while NetMHCpan 4.0 and NetMHCcons 1.1 outperform the other machine learning-based and consensus-based tools, respectively. Importantly, it should be noted that a peptide predicted with a higher binding score for a specific HLA allotype does not necessarily imply it will be immunogenic. That said, peptide-binding predictors are still very useful in that they can help to significantly reduce the large number of epitope candidates that need to be experimentally verified. Several other factors, including susceptibility to proteasome cleavage, peptide transport into the endoplasmic reticulum and T-cell receptor repertoire, also contribute to the immunogenicity of peptide antigens, and some of them can be considered by some predictors. Therefore, integrating features derived from these additional factors together with HLA-binding properties by using machine-learning algorithms may increase the prediction accuracy of immunogenic peptides. As such, we anticipate that this review and benchmarking survey will assist researchers in selecting appropriate prediction tools that best suit their purposes and provide useful guidelines for the development of improved antigen predictors in the future.

Personalized neoantigen-pulsed dendritic cell vaccines show superior immunogenicity to neoantigen-adjuvant vaccines in mouse tumor models

Development of personalized cancer vaccines based on neoantigens has become a new direction in cancer immunotherapy. Two forms of cancer vaccines have been widely studied: tumor-associated antigen (including proteins, peptides, or tumor lysates)-pulsed dendritic cell (DC) vaccines and protein- or peptide-adjuvant vaccines. However, different immune modalities may produce different therapeutic effects and immune responses when the same antigen is used. Therefore, it is necessary to choose a more effective neoantigen vaccination method. In this study, we compared the differences in immune and anti-tumor effects between neoantigen-pulsed DC vaccines and neoantigen-adjuvant vaccines using murine lung carcinoma (LL2) candidate neoantigens. The enzyme-linked immunospot (ELISPOT) assay showed that 4/6 of the neoantigen-adjuvant vaccines and 6/6 of the neoantigen-pulsed DC vaccines induced strong T-cell immune responses. Also, 2/6 of the neoantigen-adjuvant vaccines and 5/6 of the neoantigen-pulsed DC vaccines exhibited potent anti-tumor effects. The results indicated that the neoantigen-pulsed DC vaccines were superior to the neoantigen-adjuvant vaccines in both activating immune responses and inhibiting tumor growth. Our fundings provide an experimental basis for the selection of immune modalities for the use of neoantigens in individualized tumor immunotherapies.

Machine Learning for Cancer Immunotherapies Based on Epitope Recognition by T Cell Receptors

In the last years, immunotherapies have shown tremendous success as treatments for multiple types of cancer. However, there are still many obstacles to overcome in order to increase response rates and identify effective therapies for every individual patient. Since there are many possibilities to boost a patient's immune response against a tumor and not all can be covered, this review is focused on T cell receptor-mediated therapies. CD8+ T cells can detect and destroy malignant cells by binding to peptides presented on cell surfaces by MHC (major histocompatibility complex) class I molecules. CD4+ T cells can also mediate powerful immune responses but their peptide recognition by MHC class II molecules is more complex, which is why the attention has been focused on CD8+ T cells. Therapies based on the power of T cells can, on the one hand, enhance T cell recognition by introducing TCRs that preferentially direct T cells to tumor sites (so called TCR-T therapy) or through vaccination to induce T cells in vivo. On the other hand, T cell activity can be improved by immune checkpoint inhibition or other means that help create a microenvironment favorable for cytotoxic T cell activity. The manifold ways in which the immune system and cancer interact with each other require not only the use of large omics datasets from gene, to transcript, to protein, and to peptide but also make the application of machine learning methods inevitable. Currently, discovering and selecting suitable TCRs is a very costly and work intensive in vitro process. To facilitate this process and to additionally allow for highly personalized therapies that can simultaneously target multiple patient-specific antigens, especially neoepitopes, breakthrough computational methods for predicting antigen presentation and TCR binding are urgently required. Particularly, potential cross-reactivity is a major consideration since off-target toxicity can pose a major threat to patient safety. The current speed at which not only datasets grow and are made available to the public, but also at which new machine learning methods evolve, is assuring that computational approaches will be able to help to solve problems that immunotherapies are still facing.

PromPDD, a web-based tool for the prediction, deciphering and design of promiscuous peptides that bind to HLA class I molecules

Promiscuous peptides that can be presented by multiple human leukocyte antigens (HLAs) have great potential for the development of vaccines with wide population coverage. However, the current available methods for the prediction of peptides that bind to major histocompatibility complex (MHC) are mainly aimed at the rapid or mass screening of potential T cell epitopes from pathogen antigens or proteomics. The current approaches do not allow deciphering the contribution of the residue at each peptide position to the promiscuous binding ability of the peptide or obtaining guidelines for the design of promiscuous peptides. In this study, we re-evaluated and characterized four matrix-based prediction models that have been extensively used for the prediction of HLA-binding peptides and found that the prediction models generated based on the average relative binding (ARB) matrix shared a consistent and conservative threshold for all well-studied HLA class I alleles. Evaluations performed using datasets of HLA supertype-specific peptides with various cross-binding abilities and peptide mutant analogues indicated that the ARB-based binding matrices could be used to decipher and design promiscuous peptides that bind to multiple HLA molecules. A web-based tool called PromPDD was developed using ARB matrix-based models, and this tool enables the prediction, deciphering and design of promiscuous peptides that bind to multiple HLA molecules within or across HLA supertypes in a simpler and more direct manner. Furthermore, we expanded the application of PromPDD to HLA class I alleles with limited experimentally verified data by generating pan-specific matrices using a derived modular method, and 2641 HLA molecules encoded by HLA-A and HLA-B genes are available in PromPDD. PromPDD, which is freely available at http://www.immunoinformatics.net/PromPDD/, is the first tool for the deciphering and design of promiscuous peptides that bind to HLA class I molecules.

A Comprehensive Survey of Genomic Alterations in Gastric Cancer Reveals Recurrent Neoantigens as Potential Therapeutic Targets

Immunotherapy directed against cancer-specific neoantigens derived from non-silent mutants is a promising individualized strategy for cancer treatment. Neoantigens shared across patients could be used as a public resource for developing T cell-based therapy. To identify potential public neoantigens for therapy in gastric cancer (GC), 74 GC patients were enrolled in this study. Combined with the TCGA cohort and other published studies, whole exome sequencing data from 942 GC patients were used to detect somatic mutations and predict neoantigens shared by GC patients. The mutations pattern between our study and the TCGA cohort is comparable, and C > T is the most common substitution. The number of neoantigens was significantly higher in older patients (age ≥60) compared to younger patients (age <60), both in this study and the TCGA cohort. Recurrent neoantigens were found in eight genes (TP53, PIK3CA, PGM5, ERBB3, C6, TRIM49C, OR4C16, and KRAS) in this study. The neoantigen-associated mutations PIK3CA (p.H1047R) and TP53 (p.R175H) are common across several cancer types, indicating their potential usage. Overall, our study illustrates a comprehensive genomic landscape of GC and provides the recurrent neoantigens to facilitate further immunotherapy.

DeepHLApan: A Deep Learning Approach for Neoantigen Prediction Considering Both HLA-Peptide Binding and Immunogenicity

Neoantigens play important roles in cancer immunotherapy. Current methods used for neoantigen prediction focus on the binding between human leukocyte antigens (HLAs) and peptides, which is insufficient for high-confidence neoantigen prediction. In this study, we apply deep learning techniques to predict neoantigens considering both the possibility of HLA-peptide binding (binding model) and the potential immunogenicity (immunogenicity model) of the peptide-HLA complex (pHLA). The binding model achieves comparable performance with other well-acknowledged tools on the latest Immune Epitope Database (IEDB) benchmark datasets and an independent mass spectrometry (MS) dataset. The immunogenicity model could significantly improve the prediction precision of neoantigens. The further application of our method to the mutations with pre-existing T-cell responses indicating its feasibility in clinical application. DeepHLApan is freely available at https://github.com/jiujiezz/deephlapan and http://biopharm.zju.edu.cn/deephlapan.

Molecular Profiles and Metastasis Markers in Chinese Patients with Gastric Carcinoma

The goal of this work was to investigate the molecular profiles and metastasis markers in Chinese patients with gastric carcinoma (GC). In total, we performed whole exome sequencing (WES) on 74 GC patients with tumor and adjacent normal formalin-fixed, paraffin-embedded (FFPE) tissue samples. The mutation spectrum of these samples showed a high concordance with TCGA and other studies on GC. PTPRT is significantly associated with metastasis of GC, suggesting its predictive role in metastasis of GC. Patients carrying BRCA2 mutations tend not to metastasize, which may be related to their sensitivity to chemotherapy. Mutations in MACF1, CDC27, HMCN1, CDH1 and PDZD2 were moderately enriched in peritoneal metastasis (PM) samples. Furthermore, we found two genomic regions (1p36.21 and Xq26.3) were associated with PM of GC, and patients with amplification of 1p36.21 and Xq26.3 have a worse prognosis (P = 0.002, 0.01, respectively). Our analysis provides GC patients with potential markers for single and combination therapies.

The molecular landscape of synchronous colorectal cancer reveals genetic heterogeneity

Synchronous colorectal cancers (syCRCs), which present two or more lesions at diagnosis, are rare and pose a great challenge for clinical management. Although some predisposing factors associated with syCRCs have been studied with limited accession, the full repertoire of genomic events among the lesions within an individual and the causes of syCRCs remain unclear. We performed whole-exome sequencing of 40 surgical tumour samples of paired lesions from 20 patients to characterize the genetic alterations. Lesions from same patient showed distinct landscapes of somatic aberrations and shared few mutations, which suggests that they originate and develop independently, although they shared the similar genetic background. Canonical genes, such as APC, KRAS, TP53 and PIK3CA, were frequently mutated in the syCRCs, and most of them show different mutation profile compared with solitary colorectal cancer. We identified a recurrent somatic alteration (K15fs) in RPL22 in 25% of the syCRCs. Functional analysis indicated that mutated RPL22 may suppress cell apoptosis and promote the epithelial–mesenchymal transition (EMT). Potential drug targets were identified in several signalling pathways, and they present great discrepancy between lesions from the same patient. Our data show that the syCRCs within the same patient present great genetic heterogeneity, and they may be driven by distinct molecular events and develop independently. The discrepancy of potential drug targets and mutation burden in lesions from one patient provides valuable information in clinical management for patients with syCRCs.

DeepSeqPan, a novel deep convolutional neural network model for pan-specific class I HLA-peptide binding affinity prediction

Interactions between human leukocyte antigens (HLAs) and peptides play a critical role in the human immune system. Accurate computational prediction of HLA-binding peptides can be used for peptide drug discovery. Currently, the best prediction algorithms are neural network-based pan-specific models, which take advantage of the large amount of data across HLA alleles. However, current pan-specific models are all based on the pseudo sequence encoding for modeling the binding context, which is based on 34 positions identified from the HLA protein-peptide bound structures in early works. In this work, we proposed a novel deep convolutional neural network model (DCNN) for HLA-peptide binding prediction, in which the encoding of the HLA sequence and the binding context are both learned by the network itself without requiring the HLA-peptide bound structure information. Our DCNN model is also characterized by its binding context extraction layer and dual outputs with both binding affinity output and binding probability outputs. Evaluation on public benchmark datasets shows that our DeepSeqPan model without HLA structural information in training achieves state-of-the-art performance on a large number of HLA alleles with good generalization capability. Since our model only needs raw sequences from the HLA-peptide binding pairs, it can be applied to binding predictions of HLAs without structure information and can also be applied to other protein binding problems such as protein-DNA and protein-RNA bindings. The implementation code and trained models are freely available at https://github.com/pcpLiu/DeepSeqPan .

Nature of tumour rejection antigens in ovarian cancer

Major progress in the analysis of human immune responses to cancer has been made through the molecular characterization of human tumour antigens. The development of therapeutic strategies for eliciting immune-mediated rejection of tumours has accelerated due to the elucidation of the molecular basis for tumour cell recognition and destruction by immune cells. Of the various human tumour antigens defined to date in ovarian cancer, the cancer-testis (CT) family of antigens have been studied extensively preclinically and clinically because of their testis-restricted expression in normal tissues and ability to elicit robust immune responses. Recent developments in cancer sequencing technologies offer a unique opportunity to identify tumour mutations with the highest likelihood of being expressed and recognized by the immune system. Such mutations, or neoantigens, could potentially serve as specific immune targets for T-cell-mediated destruction of cancer cells. This review will highlight current work in selecting tumour rejection antigens in ovarian cancer for improving the efficacy of immunotherapy.

Predicting Antigen Presentation-What Could We Learn From a Million Peptides?

Antigen presentation lies at the heart of immune recognition of infected or malignant cells. For this reason, important efforts have been made to predict which peptides are more likely to bind and be presented by the human leukocyte antigen (HLA) complex at the surface of cells. These predictions have become even more important with the advent of next-generation sequencing technologies that enable researchers and clinicians to rapidly determine the sequences of pathogens (and their multiple variants) or identify non-synonymous genetic alterations in cancer cells. Here, we review recent advances in predicting HLA binding and antigen presentation in human cells. We argue that the very large amount of high-quality mass spectrometry data of eluted (mainly self) HLA ligands generated in the last few years provides unprecedented opportunities to improve our ability to predict antigen presentation and learn new properties of HLA molecules, as demonstrated in many recent studies of naturally presented HLA-I ligands. Although major challenges still lie on the road toward the ultimate goal of predicting immunogenicity, these experimental and computational developments will facilitate screening of putative epitopes, which may eventually help decipher the rules governing T cell recognition.

PSSMHCpan: a novel PSSM-based software for predicting class I peptide-HLA binding affinity

Predicting peptide binding affinity with human leukocyte antigen (HLA) is a crucial step in developing powerful antitumor vaccine for cancer immunotherapy. Currently available methods work quite well in predicting peptide binding affinity with HLA alleles such as HLA-A*0201, HLA-A*0101, and HLA-B*0702 in terms of sensitivity and specificity. However, quite a few types of HLA alleles that are present in the majority of human populations including HLA-A*0202, HLA-A*0203, HLA-A*6802, HLA-B*5101, HLA-B*5301, HLA-B*5401, and HLA-B*5701 still cannot be predicted with satisfactory accuracy using currently available methods. Furthermore, currently the most popularly used methods for predicting peptide binding affinity are inefficient in identifying neoantigens from a large quantity of whole genome and transcriptome sequencing data. Here we present a Position Specific Scoring Matrix (PSSM)-based software called PSSMHCpan to accurately and efficiently predict peptide binding affinity with a broad coverage of HLA class I alleles. We evaluated the performance of PSSMHCpan by analyzing 10-fold cross-validation on a training database containing 87 HLA alleles and obtained an average area under receiver operating characteristic curve (AUC) of 0.94 and accuracy (ACC) of 0.85. In an independent dataset (Peptide Database of Cancer Immunity) evaluation, PSSMHCpan is substantially better than the popularly used NetMHC-4.0, NetMHCpan-3.0, PickPocket, Nebula, and SMM with a sensitivity of 0.90, as compared to 0.74, 0.81, 0.77, 0.24, and 0.79. In addition, PSSMHCpan is more than 197 times faster than NetMHC-4.0, NetMHCpan-3.0, PickPocket, sNebula, and SMM when predicting neoantigens from 661 263 peptides from a breast tumor sample. Finally, we built a neoantigen prediction pipeline and identified 117 017 neoantigens from 467 cancer samples of various cancers from TCGA. PSSMHCpan is superior to the currently available methods in predicting peptide binding affinity with a broad coverage of HLA class I alleles.