Long Short-Term Memory to Predict 3D Amino Acids Positions in GPCR Molecular Dynamics

Abstract

G-Protein Coupled Receptors (GPCRs) are a big family of eukaryotic cell transmembrane proteins, responsible for numerous biological processes. From a practical viewpoint around 34% of the drugs approved by the US Food and Drug Administration target these receptors. They can be analyzed from their simulated molecular dynamics, including the prediction of their behavior in the presence of drugs. In this paper, the capability of Long Short-Term Memory Networks (LSTMs) are evaluated to learn and predict the molecular dynamic trajectories of a receptor. Several models were trained with the 3D position of the amino acids of the receptor considering different transformations on the position of the amino acid, such as their centers of mass, the geometric centers and the position of the α–carbon for each amino acid. The error of the prediction of the position was evaluated by the mean average error (MAE) and root-mean-square deviation (RMSD). The LSTM models show a robust performance, with results comparable to the state-of-the-art in non-dynamic 3D predictions. The best MAE and RMSD values were found for the mass center of the amino acids with 0.078 Å and 0.156 Å respectively. This work shows the potential of LSTM to predict the molecular dynamics of GPRCs.

Molecular Dynamics forecasting of transmembrane Regions in GPRCs by Recurrent Neural Networks

“…In the case of the extra-cellular regions, for both simulations, the EL1 region was identified to have the lowest prediction error, while EL2 and EL3 regions yielded a significant higher prediction error. In addition, the experimental setup was carried out with ULSM, since the best forecasting performances was obtained in [8]. This may be given because the BLSTM trains in two direction, forward and backward over the sequence.…”

“…In previous research [8], unidirectional LSTM (ULSTM) and bidirectional LSTM (BLSTM) were applied to predict the trajectories of the MD simulations of the receptor described in this study. ULSTM works by processing data in the forward direction, while BLSTM processes sequence data in both forward and backward directions with two separate hidden layers [11].…”

“…We refer to sequence lengths as nSteps-in and to length of the predicted sequences as nStepsout, and the amino acid representative data point position as center of the amino acids. Values for these parameters were chosen according to the experiments reported in [8]. These were nSteps-in = 7, nSteps-out = 1 and Center of the amino acids = "center of the mass".…”

“…In previous research [8], we carried out a preliminary study of this kind using unidirectional and bidirectional LSTM. This study revealed the usefulness of LSTM to predict MD trajectories of a GPCR receptor as a whole.…”

GPRC Molecular Dynamics Forecasting using Recurrent Neural Networks

“…We refer to sequence lengths as nSteps-in (a value of 5 in our experiments), and the amino acid representative data point position as center of mass. Values for these parameters were chosen according to the experiments reported in 37 .…”