Search for: All records

In computer-aided drug discovery, quantitative structure activity relation models are trained to predict biological activity from chemical structure. Despite the recent success of applying graph neural network to this task, important chemical information such as molecular chirality is ignored. To fill this crucial gap, we propose Molecular-Kernel Graph NeuralNetwork (MolKGNN) for molecular representation learning, which features SE(3)-/conformation invariance, chirality-awareness, and interpretability. For our MolKGNN, we first design a molecular graph convolution to capture the chemical pattern by comparing the atom's similarity with the learnable molecular kernels. Furthermore, we propagate the similarity score to capture the higher-order chemical pattern. To assess the method, we conduct a comprehensive evaluation with nine well-curated datasets spanning numerous important drug targets that feature realistic high class imbalance and it demonstrates the superiority of MolKGNN over other graph neural networks in computer-aided drug discovery. Meanwhile, the learned kernels identify patterns that agree with domain knowledge, confirming the pragmatic interpretability of this approach. Our code and supplementary material are publicly available at https://github.com/meilerlab/MolKGNN.

 

Oncogenic mutations within the epidermal growth factor receptor (EGFR) are found in 15 to 30% of all non–small-cell lung carcinomas. The term exon 19 deletion (ex19del) is collectively used to refer to more than 20 distinct genomic alterations within exon 19 that comprise the most common EGFR mutation subtype in lung cancer. Despite this heterogeneity, clinical treatment decisions are made irrespective of which EGFR ex19del variant is present within the tumor, and there is a paucity of information regarding how individual ex19del variants influence protein structure and function. Herein, we identified allele-specific functional differences among ex19del variants attributable to recurring sequence and structure motifs. We built all-atom structural models of 60 ex19del variants identified in patients and combined molecular dynamics simulations with biochemical and biophysical experiments to analyze three ex19del mutations (E746_A750, E746_S752 > V, and L747_A750 > P). We demonstrate that sequence variation in ex19del alters oncogenic cell growth, dimerization propensity, enzyme kinetics, and tyrosine kinase inhibitor (TKI) sensitivity. We show that in contrast to E746_A750 and E746_S752 > V, the L747_A750 > P variant forms highly active ligand-independent dimers. Enzyme kinetic analysis and TKI inhibition experiments suggest that E746_S752 > V and L747_A750 > P display reduced TKI sensitivity due to decreased adenosine 5′-triphosphate K m . Through these analyses, we propose an expanded framework for interpreting ex19del variants and considerations for therapeutic intervention. 

Mechanistic understanding of oncogenic variants facilitates the development and optimization of treatment strategies. We recently identified in-frame, tandem duplication ofEGFRexons 18 - 25, which causes EGFR Kinase Domain Duplication (EGFR-KDD). Here, we characterize the prevalence ofERBBfamily KDDs across multiple human cancers and evaluate the functional biochemistry of EGFR-KDD as it relates to pathogenesis and potential therapeutic intervention. We provide computational and experimental evidence that EGFR-KDD functions by forming asymmetric EGF-independent intra-molecular and EGF-dependent inter-molecular dimers. Time-resolved fluorescence microscopy and co-immunoprecipitation reveals EGFR-KDD can form ligand-dependent inter-molecular homo- and hetero-dimers/multimers. Furthermore, we show that inhibition of EGFR-KDD activity is maximally achieved by blocking both intra- and inter-molecular dimerization. Collectively, our findings define a previously unrecognized model of EGFR dimerization, providing important insights for the understanding of EGFR activation mechanisms and informing personalized treatment of patients with tumors harboring EGFR-KDD. Finally, we establishERBBKDDs as recurrent oncogenic events in multiple cancers.

 

Computational protein design has been successful in modeling fixed backbone proteins in a single conformation. However, when modeling large ensembles of flexible proteins, current methods in protein design have been insufficient. Large barriers in the energy landscape are difficult to traverse while redesigning a protein sequence, and as a result current design methods only sample a fraction of available sequence space. We propose a new computational approach that combines traditional structure-based modeling using the ROSETTA software suite with machine learning and integer linear programming to overcome limitations in the ROSETTA sampling methods. We demonstrate the effectiveness of this method, which we call BROAD, by benchmarking the performance on increasing predicted breadth of anti-HIV antibodies. We use this novel method to increase predicted breadth of naturally-occurring antibody VRC23 against a panel of 180 divergent HIV viral strains and achieve 100% predicted binding against the panel. In addition, we compare the performance of this method to state-of-the-art multistate design in ROSETTA and show that we can outperform the existing method significantly. We further demonstrate that sequences recovered by this method recover known binding motifs of broadly neutralizing anti-HIV antibodies. Finally, our approach is general and can be extended easily to other protein systems. 

Abstract Each year vast international resources are wasted on irreproducible research. The scientific community has been slow to adopt standard software engineering practices, despite the increases in high-dimensional data, complexities of workflows, and computational environments. Here we show how scientific software applications can be created in a reproducible manner when simple design goals for reproducibility are met. We describe the implementation of a test server framework and 40 scientific benchmarks, covering numerous applications in Rosetta bio-macromolecular modeling. High performance computing cluster integration allows these benchmarks to run continuously and automatically. Detailed protocol captures are useful for developers and users of Rosetta and other macromolecular modeling tools. The framework and design concepts presented here are valuable for developers and users of any type of scientific software and for the scientific community to create reproducible methods. Specific examples highlight the utility of this framework, and the comprehensive documentation illustrates the ease of adding new tests in a matter of hours. 

Computational membrane protein design is challenging due to the small number of high‐resolution structures available to elucidate the physical basis of membrane protein structure, multiple functionally important conformational states, and a limited number of high‐throughput biophysical assays to monitor function. However, structural determination of membrane proteins has made tremendous progress in the past years. Concurrently the field of soluble computational design has made impressive inroads. These developments allow us to tackle the formidable challenge of designing functional membrane proteins. Herein, Rosetta is benchmarked for membrane protein design. We evaluate strategies to cope with the often reduced quality of experimental membrane protein structures. Further, we test the usage of symmetry in design protocols, which is particularly important as many membrane proteins exist as homo‐oligomers. We compare a soluble scoring function with a scoring function optimized for membrane proteins, RosettaMembrane. Both scoring functions recovered around half of the native sequence when completely redesigning membrane proteins. However, RosettaMembrane recovered the most native‐like amino acid property composition. While leucine was overrepresented in the inner and outer‐hydrophobic regions of RosettaMembrane designs, it resulted in a native‐like surface hydrophobicity indicating that it is currently the best option for designing membrane proteins with Rosetta.