Titulo Estágio
De novo drug design with Optimized Objectives
Áreas de especialidade
Sistemas Inteligentes
Local do Estágio
Laboratory of Artificial Neural Networks (LARN)
Enquadramento
Drug discovery and development is a complicated, lengthy process, and a candidate molecule may fail due to a variety of factors, such as poor pharmacokinetics, lack of effectiveness or toxicity. Successful drug candidates necessarily represent a compromise between the numerous, sometimes conflicting features so that potential drawbacks and risks outweigh the benefits to patients. De novo drug design involves searching in an immense space of feasible, drug-like molecules to select those with the highest chance of using computational techniques to become drugs. De novo design has traditionally focused on designing molecules that meet a single objective, such as similarity to a known ligand or an interaction score, and ignored the presence of the multiple objectives required for drug-like behaviour. Methods have recently appeared in the literature that seek to design molecules that meet multiple predefined goals and thus produce candidate solutions with a higher chance of serving as viable drug lead compounds.
This proposal proposes the use of state-of-the-art Deep Learning methods to develop a computational model that accurately generates novel molecules that not only have the predicted activity against a target but also include multiple pharmacological objectives such as toxicity. We plan to introduce Multi-objective Q-learning Reinforcement Learning (MQLRL) that is a generalization of standard reinforcement learning where the scalar reward signal is extended to multiple feedback signals, in essence, one for each objective. MQLRL is the process of learning policies that optimize numerous objectives simultaneously.
This is a novel temporal difference learning algorithm that integrates the Pareto dominance relation into a reinforcement learning approach. This algorithm is a multi-policy algorithm that learns a set of Pareto dominating policies in a single run. A crucial aspect of Pareto Q-learning is the updating mechanism that bootstraps sets of Q-vectors. One of the main contributions in this proposal is a mechanism that separates the expected immediate reward vector from the set of expected future discounted reward vectors.
Objetivo
The main objective of this proposal is to develop a deep generative method combined with multi-objective reinforcement learning and analysis on real database benchmark, which contains molecules and measured biological activity data.
The main goals of this proposal are:
(i) Construct a data set for the MORL model;
(ii) Perform Data Pre-Processing, Normalization and Scaling;
(iii) Select appropriate ML algorithms for building the MQLRL model;
(iv) Perform Sampling and Model Evaluation and validate the overall model with real data.
(v) Integrate the implemented components reusable platform.
Plano de Trabalhos - Semestre 1
• Overview of drug discovery, including target identification, lead discovery, and lead optimization;
• Overview of machine learning techniques, namely Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU) and multi-objective reinforcement learning;
• Propose initial deep generative model combined with multi-objective reinforcement learning workflow and prepare the first case study
• Prepare the intermediate report.
Plano de Trabalhos - Semestre 2
• Select, and pre-process a collection of large datasets for experiments;
• Study, and select, Machine Learning (ML) algorithms for building the generative model to create valid novel drug molecules (SMILEs);
• Study and select feature selection algorithms and multi-objective reinforcement learning model machine for generating chemically feasible SMILE strings
• Analyse experimental results: e.g., study parameter values; compare performance of the reduced datasets vs. previous results, etc.;
• Prepare a research paper and the final version of the thesis.
Condições
This work will be carried out in the Laboratory of Neural Networks (LARN) of CISUC, where there will be regular supervision and feedback on behalf of the supervisors.
Familiarity with machine learning and data mining algorithms and software tools are essential. Participating students will acquire valuable knowledge and experience with model building and data science by mining massive datasets, which skills are currently in high demand for various technology employers due to the relevance to various applications.
Orientador
Maryam Abbasi,Joel P. Arrais, Bernardete Ribeiro
maryam@dei.uc.pt 📩