Sampling Conformational Ensembles of Highly Dynamic Proteins via Generative Deep Learning.

Proteins are inherently dynamic, and their conformational ensembles are functionally important in biology. Large-scale motions may govern protein structure-function relationship, and numerous transient but stable conformations of intrinsically disordered proteins (IDPs) can play a crucial role in biological function. Investigating conformational ensembles to understand regulations and disease-related aggregations of IDPs is challenging both experimentally and computationally. In this paper first an unsupervised deep learning-based model, termed Internal Coordinate Net (ICoN), is developed that learns the physical principles of conformational changes from molecular dynamics (MD) simulation data. Second, interpolating data points in the learned latent space are selected that rapidly identify novel synthetic conformations with sophisticated and large-scale sidechains and backbone arrangements. Third, with the highly dynamic amyloid-ß1-42 (Aß42) monomer, our deep learning model provided a comprehensive sampling of Aß42's conformational landscape. Analysis of these synthetic conformations revealed conformational clusters that can be used to rationalize experimental findings. Additionally, the method can identify novel conformations with important interactions in atomistic details that are not included in the training data. New synthetic conformations showed distinct sidechain rearrangements that are probed by our EPR and amino acid substitution studies. The proposed approach is highly transferable and can be used for any available data for training. The work also demonstrated the ability for deep learning to utilize learned natural atomistic motions in protein conformation sampling.

Sampling Conformational Ensembles of Highly Dynamic Proteins via Generative Deep Learning.

Proteins are inherently dynamic, and their conformational ensembles are functionally important in biology. Large-scale motions may govern protein structure-function relationship, and numerous transient but stable conformations of intrinsically disordered proteins (IDPs) can play a crucial role in biological function. Investigating conformational ensembles to understand regulations and disease-related aggregations of IDPs is challenging both experimentally and computationally. In this paper we first introduced an unsupervised deep learning-based model, termed Internal Coordinate Net (ICoN), which learns the physical principles of conformational changes from molecular dynamics (MD) simulation data. Second, we selected interpolating data points in the learned latent space that rapidly identify novel synthetic conformations with sophisticated and large-scale sidechains and backbone arrangements. Third, with the highly dynamic amyloid-ß 1-42 (Aß42) monomer, our deep learning model provided a comprehensive sampling of Aß42's conformational landscape. Analysis of these synthetic conformations revealed conformational clusters that can be used to rationalize experimental findings. Additionally, the method can identify novel conformations with important interactions in atomistic details that are not included in the training data. New synthetic conformations showed distinct sidechain rearrangements that are probed by our EPR and amino acid substitution studies. This approach is highly transferable and can be used for any available data for training. The work also demonstrated the ability for deep learning to utilize learned natural atomistic motions in protein conformation sampling.

Selective recognition and discrimination of single isomeric changes in peptide strands with a host : guest sensing array.

An indirect competitive binding mechanism can be exploited to allow a combination of cationic fluorophores and water-soluble synthetic receptors to selectively recognize and discriminate peptide strands containing a single isomeric residue in the backbone. Peptide isomerization occurs in long-lived proteins and has been linked with diseases such as Alzheimer's, cataracts and cancer, so isomers are valuable yet underexplored targets for selective recognition. Planar cationic fluorophores can selectively bind hydrophobic, Trp-containing peptide strands in solution, and when paired with receptors that provide a competitive host for the fluorophore, can form a differential sensing array that enables selective discrimination of peptide isomers. Residue variations such as D- and L-Asp, D- and L-isoAsp, D-Ser and D-Glu can all be recognized, simply by their effects on the folded structure of the flexible peptide. Molecular dynamics simulations were applied to determine the most favorable conformation of the peptide : fluorophore conjugate, indicating that favorable π-stacking with internal tryptophan residues in a folded binding pocket enables micromolar binding affinity.