De-novo computational design of peptides for the specific and selective recognition of small molecules

Bibliographic Details
Main Author: Pereira, José Manuel Santos
Publication Date: 2023
Language: eng
Source: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
Download full: http://hdl.handle.net/10773/37884
Summary: The main objective of this PhD thesis was the computational design of peptides for the specific and selective recognition of small molecules. In the last decades, the field of computational protein design has seen massive developments both at the hardware and software level. Some of the most commonly used tools, however, have lagged behind in comparison with the expectations of a modern protein design software: speed, simplicity and a wide range of features. In that regard, and on the scope of the present thesis, a new solution for the computational design of proteins was envisioned, named ProtoSyn.jl. This tool, developed in a model programming language (Julia), intends to facilitate the manipulation and simulation of molecular systems, with a clear emphasis on the rational design of proteins. The proposed work was divided into 4 stages: (i) The definition of a base architecture for ProtoSyn.jl, including data structures representative of molecular systems and all the code related to the simulation of said systems; (ii) the development of numerical optimization algorithms based on SIMD, GPU and distributed computing technologies for the calculation of non-bonded atomic interactions in molecular systems; (iii) The development and parametrization of energy function components for the iterative evaluation of molecular structures; (iv) The application of the developed program in practical simulation use-cases, namely a) sidechain repackaging, b) protein folding prediction, c) amino acid sequence recovery, d) small molecule docking and, finally, e) rational design of peptides for the recognition of small molecules. In short, this PhD thesis was based on a multidisciplinary approach, distributed between the fields of computational chemistry and computer sciences. Its main result is ProtoSyn.jl: a novel API for molecular manipulation and simulation, constituting a modern solution that allows and facilitates the composition of arbitrarily complex simulation protocols in a simple and modular fashion. Furthermore, several practical applications of the program are presented, culminating on the definition of a protein design candidate library for the specific recognition of a small molecule on the surface of a peptide.
id RCAP_ba8130ec6c0fccebfb7dfc96ce7dc696
oai_identifier_str oai:ria.ua.pt:10773/37884
network_acronym_str RCAP
network_name_str Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
repository_id_str https://opendoar.ac.uk/repository/7160
spelling De-novo computational design of peptides for the specific and selective recognition of small moleculesComputational protein designMolecular dockingMolecular modellingMolecular simulationSoftware developmentJulia (programming language)The main objective of this PhD thesis was the computational design of peptides for the specific and selective recognition of small molecules. In the last decades, the field of computational protein design has seen massive developments both at the hardware and software level. Some of the most commonly used tools, however, have lagged behind in comparison with the expectations of a modern protein design software: speed, simplicity and a wide range of features. In that regard, and on the scope of the present thesis, a new solution for the computational design of proteins was envisioned, named ProtoSyn.jl. This tool, developed in a model programming language (Julia), intends to facilitate the manipulation and simulation of molecular systems, with a clear emphasis on the rational design of proteins. The proposed work was divided into 4 stages: (i) The definition of a base architecture for ProtoSyn.jl, including data structures representative of molecular systems and all the code related to the simulation of said systems; (ii) the development of numerical optimization algorithms based on SIMD, GPU and distributed computing technologies for the calculation of non-bonded atomic interactions in molecular systems; (iii) The development and parametrization of energy function components for the iterative evaluation of molecular structures; (iv) The application of the developed program in practical simulation use-cases, namely a) sidechain repackaging, b) protein folding prediction, c) amino acid sequence recovery, d) small molecule docking and, finally, e) rational design of peptides for the recognition of small molecules. In short, this PhD thesis was based on a multidisciplinary approach, distributed between the fields of computational chemistry and computer sciences. Its main result is ProtoSyn.jl: a novel API for molecular manipulation and simulation, constituting a modern solution that allows and facilitates the composition of arbitrarily complex simulation protocols in a simple and modular fashion. Furthermore, several practical applications of the program are presented, culminating on the definition of a protein design candidate library for the specific recognition of a small molecule on the surface of a peptide.Esta tese de doutoramento teve como principal objetivo o design (desenho) computacional de péptidos para o reconhecimento específico e seletivo de pequenas moléculas. Nas últimas décadas, o campo científico do desenho computacional de péptidos tem vindo a sofrer um enorme desenvolvimento, com melhorias tanto a nível de hardware como de software. Algumas das ferramentas mais utilizadas têm, no entanto, sentido dificuldades em corresponder às expetativas colocadas num programa de desenho proteico moderno: rapidez, simplicidade e um leque alargado de funcionalidades. Nesse sentido, e no âmbito desta tese, procurou-se desenvolver uma nova solução computacional para o desenho racional de péptidos, intitulada de ProtoSyn.jl. Esta ferramenta, desenvolvida numa linguagem de programação moderna (Julia), procura facilitar a manipulação e simulação de sistemas moleculares, com um claro foco no campo do desenho racional de proteínas. O trabalho proposto foi dividido em 4 partes: (i) A definição de uma arquitetura base para o ProtoSyn.jl, incluindo estruturas de dados representativas de sistemas moleculares e todo o código relacionado com a simulação computacional desses mesmos sistemas; (ii) O desenvolvimento de algoritmos de otimização numérica baseados em tecnologias como SIMD, GPU e computação distribuída para o cálculo de interações atómicas não-covalentes em sistemas moleculares; (iii) O desenvolvimento e parametrização de componentes de função de energia para a avaliação iterativa de estruturas moleculares; (iv) A aplicação do programa desenvolvido em exemplos práticos de simulação, nomeadamente a) re-empacotamento de cadeias laterais, b) previsão do enrolamento proteico, c) recuperação de sequências de aminoácidos, d) identificação de pontos de ancoragem para pequenas moléculas e, finalmente, e) desenho racional de péptidos para a identificação de pequenas moléculas. Em suma, esta tese de doutoramento baseou-se numa abordagem multidisciplinar bipartida entre os campos da química computacional e ciência dos computadores, tendo-se como principal resultado o ProtoSyn.jl: uma nova API para a manipulação e simulação molecular, constituindo uma solução moderna que permite e facilita a composição de protocolos de simulação molecular arbitrariamente complexos de forma simples e modular. Para além disso, são ainda apresentadas várias aplicações práticas do programa, culminando na definição de uma biblioteca de candidatos de design proteico para o reconhecimento específico de uma pequena molécula na superfície de um péptido.2023-05-25T14:03:44Z2023-04-03T00:00:00Z2023-04-03doctoral thesisinfo:eu-repo/semantics/publishedVersionapplication/pdfhttp://hdl.handle.net/10773/37884engPereira, José Manuel Santosinfo:eu-repo/semantics/openAccessreponame:Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)instname:FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologiainstacron:RCAAP2024-05-06T04:45:31Zoai:ria.ua.pt:10773/37884Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireinfo@rcaap.ptopendoar:https://opendoar.ac.uk/repository/71602025-05-28T14:19:16.067650Repositórios Científicos de Acesso Aberto de Portugal (RCAAP) - FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologiafalse
dc.title.none.fl_str_mv De-novo computational design of peptides for the specific and selective recognition of small molecules
title De-novo computational design of peptides for the specific and selective recognition of small molecules
spellingShingle De-novo computational design of peptides for the specific and selective recognition of small molecules
Pereira, José Manuel Santos
Computational protein design
Molecular docking
Molecular modelling
Molecular simulation
Software development
Julia (programming language)
title_short De-novo computational design of peptides for the specific and selective recognition of small molecules
title_full De-novo computational design of peptides for the specific and selective recognition of small molecules
title_fullStr De-novo computational design of peptides for the specific and selective recognition of small molecules
title_full_unstemmed De-novo computational design of peptides for the specific and selective recognition of small molecules
title_sort De-novo computational design of peptides for the specific and selective recognition of small molecules
author Pereira, José Manuel Santos
author_facet Pereira, José Manuel Santos
author_role author
dc.contributor.author.fl_str_mv Pereira, José Manuel Santos
dc.subject.por.fl_str_mv Computational protein design
Molecular docking
Molecular modelling
Molecular simulation
Software development
Julia (programming language)
topic Computational protein design
Molecular docking
Molecular modelling
Molecular simulation
Software development
Julia (programming language)
description The main objective of this PhD thesis was the computational design of peptides for the specific and selective recognition of small molecules. In the last decades, the field of computational protein design has seen massive developments both at the hardware and software level. Some of the most commonly used tools, however, have lagged behind in comparison with the expectations of a modern protein design software: speed, simplicity and a wide range of features. In that regard, and on the scope of the present thesis, a new solution for the computational design of proteins was envisioned, named ProtoSyn.jl. This tool, developed in a model programming language (Julia), intends to facilitate the manipulation and simulation of molecular systems, with a clear emphasis on the rational design of proteins. The proposed work was divided into 4 stages: (i) The definition of a base architecture for ProtoSyn.jl, including data structures representative of molecular systems and all the code related to the simulation of said systems; (ii) the development of numerical optimization algorithms based on SIMD, GPU and distributed computing technologies for the calculation of non-bonded atomic interactions in molecular systems; (iii) The development and parametrization of energy function components for the iterative evaluation of molecular structures; (iv) The application of the developed program in practical simulation use-cases, namely a) sidechain repackaging, b) protein folding prediction, c) amino acid sequence recovery, d) small molecule docking and, finally, e) rational design of peptides for the recognition of small molecules. In short, this PhD thesis was based on a multidisciplinary approach, distributed between the fields of computational chemistry and computer sciences. Its main result is ProtoSyn.jl: a novel API for molecular manipulation and simulation, constituting a modern solution that allows and facilitates the composition of arbitrarily complex simulation protocols in a simple and modular fashion. Furthermore, several practical applications of the program are presented, culminating on the definition of a protein design candidate library for the specific recognition of a small molecule on the surface of a peptide.
publishDate 2023
dc.date.none.fl_str_mv 2023-05-25T14:03:44Z
2023-04-03T00:00:00Z
2023-04-03
dc.type.driver.fl_str_mv doctoral thesis
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
status_str publishedVersion
dc.identifier.uri.fl_str_mv http://hdl.handle.net/10773/37884
url http://hdl.handle.net/10773/37884
dc.language.iso.fl_str_mv eng
language eng
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.source.none.fl_str_mv reponame:Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
instname:FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologia
instacron:RCAAP
instname_str FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologia
instacron_str RCAAP
institution RCAAP
reponame_str Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
collection Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
repository.name.fl_str_mv Repositórios Científicos de Acesso Aberto de Portugal (RCAAP) - FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologia
repository.mail.fl_str_mv info@rcaap.pt
_version_ 1833594497521418240