Login / Signup

Determination of Hydrophobic Lengths of Membrane Proteins with the HDGB Implicit Membrane Model.

Bercem DutagaciMichael Feig
Published in: Journal of chemical information and modeling (2017)
A protocol for predicting the hydrophobic length of membrane proteins using the heterogeneous dielectric generalized Born (HDGB) implicit membrane model is presented. The method involves optimal positioning in the membrane and identification of lipid-facing and inward-facing residues, followed by energy optimization of the implicit membrane model to obtain the hydrophobic length from the optimal membrane width. The latest HDGB version 3 (HDGBv3) and HDGB van der Waals (HDGBvdW) models were applied to a test set containing 15 proteins (seven β-barrel and eight α-helical proteins), for which matching membrane widths are available from experiment, and an additional set contains ten α-helical and ten β-barrel proteins without any experimental data. The results with the HDGB model compare favorably with predictions from methods used in the Orientations of Proteins in Membranes (OPM) and Protein Data Bank of Transmembrane Proteins (PDB-TM) databases.
Keyphrases
  • randomized controlled trial
  • electronic health record
  • big data
  • ionic liquid
  • mass spectrometry
  • machine learning
  • deep learning
  • high resolution
  • liquid chromatography
  • low birth weight
  • data analysis