Vaccination demonstrated to be the best instrument to prevent infectious diseases. In recent years, reverse vaccinology (RV) has emerged as an innovative approach, revolutionizing vaccine discovery by bioinformatics. NERVE (New Enhanced Reverse Vaccinology Environment), the first published software environment for RV, was developed by former students from our team and it was followed by several RV pipelines for vaccine discovery against bacteria and viruses. However, there is still a need for vaccines against many eukaryotic pathogens, and indeed there is only one RV software available for eukaryotic organisms, called ’Vacceed’. This thesis aims to present the state-of-the-art in reverse vaccinology and introduce eNERVE, a novel NERVE pipeline designed for discovering protein and epitope-based vaccines against eukaryotic pathogens. eNERVE is a dynamic, stand-alone, and high-throughput software that operates on Linux/Unix systems. It accepts proteome FASTA files as input and generates a final list of probable Protein Vaccine Candidates (PVCs), each described by a saliency score. The software’s strength lies in its utilization of alignment-free methods, primarily based on machine learning and deep learning models that enable the reduction of biases and speed up the analysis process. By using proteomes instead of entire genomes, eNERVE provides insights into the potential set of expressed proteins at different developmental stages of the organism and makes it possible to analyze only the proteins that are actually synthesized. The pipeline comprises 10 modules, each producing a series of information about every protein, saved as it instance attributes for the homonym class. Data passes into various modules selected by the user (localization, adhesin, autoimmunity, annotation, epitope, select) and the final output is a CSV file containing a set of columns indicating the score, localization, epitopes, and other relevant protein properties. eNERVE is available as a stand-alone software executable by “Docker” (a platform designed to help users execute modern applications without the need to manually install dependencies). In summary, the combination of reverse vaccinology and eNERVE software offers promising avenues for accelerating vaccine development against eukaryotic pathogens. The software’s innovative features, alignment-free methods, and proteome-based approach contribute to the identification and scoring of PVCs for the next, experimental validation. Docker makes eNERVE usage suitable for researchers with a poor understanding of Unix-based systems and the command line.

Vaccination demonstrated to be the best instrument to prevent infectious diseases. In recent years, reverse vaccinology (RV) has emerged as an innovative approach, revolutionizing vaccine discovery by bioinformatics. NERVE (New Enhanced Reverse Vaccinology Environment), the first published software environment for RV, was developed by former students from our team and it was followed by several RV pipelines for vaccine discovery against bacteria and viruses. However, there is still a need for vaccines against many eukaryotic pathogens, and indeed there is only one RV software available for eukaryotic organisms, called ’Vacceed’. This thesis aims to present the state-of-the-art in reverse vaccinology and introduce eNERVE, a novel NERVE pipeline designed for discovering protein and epitope-based vaccines against eukaryotic pathogens. eNERVE is a dynamic, stand-alone, and high-throughput software that operates on Linux/Unix systems. It accepts proteome FASTA files as input and generates a final list of probable Protein Vaccine Candidates (PVCs), each described by a saliency score. The software’s strength lies in its utilization of alignment-free methods, primarily based on machine learning and deep learning models that enable the reduction of biases and speed up the analysis process. By using proteomes instead of entire genomes, eNERVE provides insights into the potential set of expressed proteins at different developmental stages of the organism and makes it possible to analyze only the proteins that are actually synthesized. The pipeline comprises 10 modules, each producing a series of information about every protein, saved as it instance attributes for the homonym class. Data passes into various modules selected by the user (localization, adhesin, autoimmunity, annotation, epitope, select) and the final output is a CSV file containing a set of columns indicating the score, localization, epitopes, and other relevant protein properties. eNERVE is available as a stand-alone software executable by “Docker” (a platform designed to help users execute modern applications without the need to manually install dependencies). In summary, the combination of reverse vaccinology and eNERVE software offers promising avenues for accelerating vaccine development against eukaryotic pathogens. The software’s innovative features, alignment-free methods, and proteome-based approach contribute to the identification and scoring of PVCs for the next, experimental validation. Docker makes eNERVE usage suitable for researchers with a poor understanding of Unix-based systems and the command line.

Machine Learning for Reverse Vaccinology: development of a bioinformatic pipeline for vaccine candidates discovery in Eukaryotic pathogen proteomes

PATANÈ, FRANCESCO
2022/2023

Abstract

Vaccination demonstrated to be the best instrument to prevent infectious diseases. In recent years, reverse vaccinology (RV) has emerged as an innovative approach, revolutionizing vaccine discovery by bioinformatics. NERVE (New Enhanced Reverse Vaccinology Environment), the first published software environment for RV, was developed by former students from our team and it was followed by several RV pipelines for vaccine discovery against bacteria and viruses. However, there is still a need for vaccines against many eukaryotic pathogens, and indeed there is only one RV software available for eukaryotic organisms, called ’Vacceed’. This thesis aims to present the state-of-the-art in reverse vaccinology and introduce eNERVE, a novel NERVE pipeline designed for discovering protein and epitope-based vaccines against eukaryotic pathogens. eNERVE is a dynamic, stand-alone, and high-throughput software that operates on Linux/Unix systems. It accepts proteome FASTA files as input and generates a final list of probable Protein Vaccine Candidates (PVCs), each described by a saliency score. The software’s strength lies in its utilization of alignment-free methods, primarily based on machine learning and deep learning models that enable the reduction of biases and speed up the analysis process. By using proteomes instead of entire genomes, eNERVE provides insights into the potential set of expressed proteins at different developmental stages of the organism and makes it possible to analyze only the proteins that are actually synthesized. The pipeline comprises 10 modules, each producing a series of information about every protein, saved as it instance attributes for the homonym class. Data passes into various modules selected by the user (localization, adhesin, autoimmunity, annotation, epitope, select) and the final output is a CSV file containing a set of columns indicating the score, localization, epitopes, and other relevant protein properties. eNERVE is available as a stand-alone software executable by “Docker” (a platform designed to help users execute modern applications without the need to manually install dependencies). In summary, the combination of reverse vaccinology and eNERVE software offers promising avenues for accelerating vaccine development against eukaryotic pathogens. The software’s innovative features, alignment-free methods, and proteome-based approach contribute to the identification and scoring of PVCs for the next, experimental validation. Docker makes eNERVE usage suitable for researchers with a poor understanding of Unix-based systems and the command line.
2022
Machine Learning for Reverse Vaccinology: development of a bioinformatic pipeline for vaccine candidates discovery in Eukaryotic pathogen proteomes
Vaccination demonstrated to be the best instrument to prevent infectious diseases. In recent years, reverse vaccinology (RV) has emerged as an innovative approach, revolutionizing vaccine discovery by bioinformatics. NERVE (New Enhanced Reverse Vaccinology Environment), the first published software environment for RV, was developed by former students from our team and it was followed by several RV pipelines for vaccine discovery against bacteria and viruses. However, there is still a need for vaccines against many eukaryotic pathogens, and indeed there is only one RV software available for eukaryotic organisms, called ’Vacceed’. This thesis aims to present the state-of-the-art in reverse vaccinology and introduce eNERVE, a novel NERVE pipeline designed for discovering protein and epitope-based vaccines against eukaryotic pathogens. eNERVE is a dynamic, stand-alone, and high-throughput software that operates on Linux/Unix systems. It accepts proteome FASTA files as input and generates a final list of probable Protein Vaccine Candidates (PVCs), each described by a saliency score. The software’s strength lies in its utilization of alignment-free methods, primarily based on machine learning and deep learning models that enable the reduction of biases and speed up the analysis process. By using proteomes instead of entire genomes, eNERVE provides insights into the potential set of expressed proteins at different developmental stages of the organism and makes it possible to analyze only the proteins that are actually synthesized. The pipeline comprises 10 modules, each producing a series of information about every protein, saved as it instance attributes for the homonym class. Data passes into various modules selected by the user (localization, adhesin, autoimmunity, annotation, epitope, select) and the final output is a CSV file containing a set of columns indicating the score, localization, epitopes, and other relevant protein properties. eNERVE is available as a stand-alone software executable by “Docker” (a platform designed to help users execute modern applications without the need to manually install dependencies). In summary, the combination of reverse vaccinology and eNERVE software offers promising avenues for accelerating vaccine development against eukaryotic pathogens. The software’s innovative features, alignment-free methods, and proteome-based approach contribute to the identification and scoring of PVCs for the next, experimental validation. Docker makes eNERVE usage suitable for researchers with a poor understanding of Unix-based systems and the command line.
Machine learning
Bioinformatics
Vaccinology
Eukaryotic pathogens
File in questo prodotto:
Non ci sono file associati a questo prodotto.

The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12608/51806