3D-Scaffold: A Deep Learning Framework to Generate 3D Coordinates of Drug-like Molecules with Desired Scaffolds.

The prerequisite of therapeutic drug design and discovery is to identify novel molecules and developing lead candidates with desired biophysical and biochemical properties. Deep generative models have demonstrated their ability to find such molecules by exploring a huge chemical space efficiently. An effective way to generate new molecules with desired target properties is by constraining the critical fucntional groups or the core scaffolds in the generation process. To this end, we developed a domain aware generative framework called 3D-Scaffold that takes 3D coordinates of the desired scaffold as an input and generates 3D coordinates of novel therapeutic candidates as an output while always preserving the desired scaffolds in generated structures. We demonstrated that our framework generates predominantly valid, unique, novel, and experimentally synthesizable molecules that have drug-like properties similar to the molecules in the training set. Using domain specific data sets, we generate covalent and noncovalent antiviral inhibitors targeting viral proteins. To measure the success of our framework in generating therapeutic candidates, generated structures were subjected to high throughput virtual screening via docking simulations, which shows favorable interaction against SARS-CoV-2 main protease (Mpro) and nonstructural protein endoribonuclease (NSP15) targets. Most importantly, our deep learning model performs well with relatively small 3D structural training data and quickly learns to generalize to new scaffolds, highlighting its potential application to other domains for generating target specific candidates.

Clinical, Serological, and Histopathological Similarities Between Severe COVID-19 and Acute Exacerbation of Connective Tissue Disease-Associated Interstitial Lung Disease (CTD-ILD).

Background and Objectives: Understanding the pathophysiology of respiratory failure in coronavirus disease 2019 (COVID-19) is indispensable for development of therapeutic strategies. Since we observed similarities between COVID-19 and interstitial lung disease in connective tissue disease (CTD-ILD), we investigated features of autoimmunity in SARS-CoV-2-associated respiratory failure. Methods: We prospectively enrolled 22 patients with RT-PCR-confirmed SARS-CoV-2 infection and 10 patients with non-COVID-19-associated pneumonia. Full laboratory testing was performed including autoantibody (AAB; ANA/ENA) screening using indirect immunofluorescence and immunoblot. Fifteen COVID-19 patients underwent high-resolution computed tomography. Transbronchial biopsies/autopsy tissue samples for histopathology and ultrastructural analyses were obtained from 4/3 cases, respectively. Results: Thirteen (59.1%) patients developed acute respiratory distress syndrome (ARDS), and five patients (22.7%) died from the disease. ANA titers ≥1:320 and/or positive ENA immunoblots were detected in 11/13 (84.6%) COVID-19 patients with ARDS, in 1/9 (11.1%) COVID-19 patients without ARDS (p = 0.002) and in 4/10 (40%) patients with non-COVID-19-associated pneumonias (p = 0.039). Detection of AABs was significantly associated with a need for intensive care treatment (83.3 vs. 10%; p = 0.002) and occurrence of severe complications (75 vs. 20%, p = 0.03). Radiological and histopathological findings were highly heterogeneous including patterns reminiscent of exacerbating CTD-ILD, while ultrastructural analyses revealed interstitial thickening, fibroblast activation, and deposition of collagen fibrils. Conclusions: We are the first to report overlapping clinical, serological, and imaging features between severe COVID-19 and acute exacerbation of CTD-ILD. Our findings indicate that autoimmune mechanisms determine both clinical course and long-term sequelae after SARS-CoV-2 infection, and the presence of autoantibodies might predict adverse clinical course in COVID-19 patients.