Timezone: »

 
Structured Q-learning For Antibody Design
Alexander Cowen-Rivers · Philip John Gorinski · aivar sootla · Asif Khan · Jun WANG · Jan Peters · Haitham Bou Ammar

@
in Reinforcement Learning for Real Life (RL4RealLife) Workshop »
Optimizing combinatorial structures is core to many real-world problems, such as those encountered in life sciences. For example, one of the crucial steps involved in antibody design is to find an arrangement of amino acids in a protein sequence that improves its binding with a pathogen. Combinatorial optimization of antibodies is difficult due to extremely large search spaces and non-linear objectives. Even for modest antibody design problems, where proteins have a sequence length of eleven, we are faced with searching over $2.05 \times 10^{14}$ structures. Applying traditional Reinforcement Learning algorithms such as Q-learning to combinatorial optimization results in poor performance. We propose Structured Q-learning (SQL), an extension of Q-learning that incorporates structural priors for combinatorial optimization. Using a molecular docking simulator, we demonstrate that SQL finds high binding energy sequences and performs favourably against baselines on eight challenging antibody design tasks, including designing antibodies for SARS-COV.

Author Information

Alexander Cowen-Rivers (University College London)
Philip John Gorinski (Huawei Noah's Ark Lab)
aivar sootla (HAUWEI )
Asif Khan (University of Edinburgh)
Jun WANG (UCL)
Jan Peters (TU Darmstadt)
Haitham Bou Ammar (Huawei)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors