Solvent and Lipid Accessibility Prediction as a Basis for Model Quality Assessment in Soluble and Membrane Proteins

Abstract

On-going efforts to improve protein structure prediction stimulate the development of scoring functions and methods for model quality assessment (MQA) that can be used to rank and select the best protein models for further refinement. In this work, sequence-based prediction of relative solvent accessibility (RSA) is employed as a basis for a simple MQA method for soluble proteins, and subsequently extended to the much less explored case of (alpha-helical) membrane proteins. In analogy to soluble proteins, the level of exposure to the lipid of amino acid residues in transmembrane (TM) domains is captured in terms of the relative lipid accessibility (RLA), which is predicted from sequence using low-complexity Support Vector Regression models. On an independent set of 23 TM proteins, the new SVR-based predictor yields correlation coefficient (CC) of 0.56 between the predicted and observed RLA profiles, as opposed to CC of 0.13 for a baseline predictor that utilizes TMLIP2H empirical lipophilicity scale (with standard deviations of about 0.15). A simple MQA approach is then defined by ranking models of membrane proteins in terms of consistency between predicted and observed RLA profiles, as a measure of similarity to the native structure. The new method does not require a set of decoy models to optimize parameters, circumventing current limitations in this regard. Several different sets of models, including those generated by fragment based folding simulations, and decoys obtained by swapping TM helices to mimic errors in template based assignment, are used to assess the new approach. Predicted RLA profiles can be used to successfully discriminate near native models from non-native decoys in most cases, significantly improving the separation of correct and incorrectly folded models compared to a simple baseline approach that utilizes TMLIP2H. As suggested by the robust performance of a simple MQA method for soluble proteins that utilizes more accurate RSA predictions, further significant improvements are likely to be achieved. The steady growth in the number of resolved membrane protein structures is expected to yield enhanced RLA predictions, facilitating further efforts to improve de novo and template based prediction of membrane protein structure.

Transmembrane Protein Alignment and Fold Recognition Based on Predicted Topology

“…The average ASA of 20 amino acids in TMPs are significantly different from that of soluble proteins, even in non-TM segments [65]; hence, ASA predictors of soluble proteins are not applicable to TMPs. However, some studies on predicting ASA specifically for TMPs [65]–[67] have been developed, which showed significantly improved accuracy of ASA prediction in TM segments. We used one of these methods MPRAP [67] to predict ASA for both targets and templates, which separates different segments of TMP and predicts the entire TMP sequence without using its topology structures as input.…”

“…However, some studies on predicting ASA specifically for TMPs [65]–[67] have been developed, which showed significantly improved accuracy of ASA prediction in TM segments. We used one of these methods MPRAP [67] to predict ASA for both targets and templates, which separates different segments of TMP and predicts the entire TMP sequence without using its topology structures as input. To reduce the impact of prediction errors in the alignment, both the target and template used predicted ASA to construct profiles.…”

Folding Membrane Proteins by Deep Transfer Learning

“…By adding MP-specific features such as lipid accessibility (Phatak et al, 2011) and topology structure (Tsirigos et al, 2015) to our deep learning models, we can improve contact accuracy by about 1%. This might be because our deep learning models have already implicitly learned them.…”

OMPcontact: An Outer Membrane Protein Inter-Barrel Residue Contact Prediction Method

“…Therefore, accessible surface area (ASA) is much different for OMPs to be predicted. We used a recent published method MPRAP (Phatak et al, 2011) to generate the ASA for all the residues in each sequence in our dataset. The predicted ASA were directly used in contact prediction as an element in the input feature vector.…”