Unit:
Κατεύθυνση ΒιοπληροφορικήΠληροφορική
Supervisors info:
Martin Reczko, Ειδικός Λειτουργικός Επιστήμονας Α' Ερευνητικού Κέντρου Βιοϊατρικών Επιστημών 'Αλέξανδρος Φλέμινγκ'
Ηλίας Μανωλάκος, Καθηγητής του τμήματος ΠΛηροφορικής και Τηλεπικοινωνιών, Εθνικό και Καποδιστριακό Πανεπιστήμιο Αθηνών
Original Title:
A Deep Reinforcement Learning Neural Network Folding Proteins
Translated title:
A Deep Reinforcement Learning Neural Network Folding Proteins
Summary:
Despite considerable progress, ab initio protein structure prediction remains unoptimised. A crowdsourcing approach is the online puzzle video game Foldit [1], that provided several useful results that matched or even outperformed algorithmically computed solutions [2]. Using Foldit, the WeFold [3] crowd had several successful participations in the Critical Assessment of Techniques for Protein Structure Prediction. Based on the recent Foldit standalone version [4], we trained a deep reinforcement neural network called DeepFoldit to improve the score assigned to an unfolded protein, using the Q-learning method [5] with experience replay. The thesis is focused on model improvement through hyperparameter tuning. We examined various implementations by examining different model architectures and changing hyperparameter values to improve the accuracy of the model. The new model’s hyper-parameters also improved its ability to generalize. Initial results, from the latest implementation, show that given a set of small unfolded training proteins, DeepFoldit learns action sequences that improve the score both on the training set and on novel test proteins. This is important as improving the game score means obtaining a better folding, taking us one step closer to the solution. Our approach combines the intuitive user interface of Foldit with the efficiency of deep reinforcement learning.
Main subject category:
Technology - Computer science
Keywords:
ab initio protein structure prediction, Reinforcement Learning, Deep Learning, Convolution Neural Networks, Q-learning with experience replay, Foldit
Number of references:
154
