Protein structure prediction with a language model improves accuracy for orphan and designed proteins.

Echocardiography is essential to cardiology. However, the need for human interpretation has limited echocardiography's full potential for precision medicine. Deep learning is an emerging tool for analyzing images but has not yet been widely applied to echocardiograms, partly due to their complex multi-view format. The essential first step toward comprehensive computer-assisted echocardiographic interpretation is determining whether computers can learn to recognize these views. We trained a convolutional neural network to simultaneously classify 15 standard views (12 video, 3 still), based on labeled still images and videos from 267 transthoracic echocardiograms that captured a range of real-world clinical variation. Our model classified among 12 video views with 97.8% overall test accuracy without overfitting. Even on single low-resolution images, accuracy among 15 views was 91.7% vs. 70.2-84.0% for board-certified echocardiographers. Data visualization experiments showed that the model recognizes similarities among related views and classifies using clinically relevant image features. Our results provide a foundation for artificial intelligence-assisted echocardiographic interpretation.

Bacterial adhesion to collagen, the most abundant protein in humans, is a critical step in the initiation and persistence of numerous bacterial infections. In this study, we explore the collagen binding mechanism of the multi-modular cell wall anchored collagen adhesin (CNA) in Staphylococcus aureus and examine how applied mechanical forces can modulate adhesion ability. The common structural-functional elements and domain organization of CNA are present across over 50 genera of bacteria. Through the use of molecular dynamics models and normal mode analysis, we shed light on the CNA’s structural and conformational dynamics and its interactions with collagen that lead to collagen binding. Our results suggest that the linker region, CNA_165-173, acts as a hinge exhibiting bending, extensional, and torsional modes of structural flexibility and its residues are key in the interaction of the CNA-collagen complex. Steered molecular dynamics simulations were conducted with umbrella sampling. During the course of these simulations, the ‘locking’ latch from the CNA N2 domain was dissociated from its groove in the CNA N1 domain, implying the importance of the latch for effective ligand binding. Finally, we observed that the binding efficiency of the CNA N1-N2 domains to collagen decreases greatly with increasing tensile force application to the collagen peptides. Thus, CNA and similar adhesins might preferentially bind to sites in which collagen fibers are cleaved, such as in wounded, injured, or inflamed tissues, or in which the collagenous tissue is less mature. As alternative techniques for control of bacterial infection are in-demand due to the rise of bacterial antibiotic resistance, results from our computational studies with respect to the mechanoregulation of the collagen binding site may inspire new therapeutics and engineering solutions by mechanically preventing colonization and/or further pathogenesis.

Diagnosis of endoleak following endovascular aortic repair (EVAR) relies on manual review of multi-slice CT angiography (CTA) by physicians which is a tedious and time-consuming process that is susceptible to error. We evaluate the use of a deep neural network for the detection of endoleak on CTA for post-EVAR patients using a novel data efficient training approach. 50 CTAs and 20 CTAs with and without endoleak respectively were identified based on gold standard interpretation by a cardiovascular subspecialty radiologist. The Endoleak Augmentor, a custom designed augmentation method, provided robust training for the machine learning (ML) model. Predicted segmentation maps underwent post-processing to determine the presence of endoleak. The model was tested against 3 blinded general radiologists and 1 blinded subspecialist using a held-out subset (10 positive endoleak CTAs, 10 control CTAs). Model accuracy, precision and recall for endoleak diagnosis were 95%, 90% and 100% relative to reference subspecialist interpretation (AUC = 0.99). Accuracy, precision and recall was 70/70/70% for generalist1, 50/50/90% for generalist2, and 90/83/100% for generalist3. The blinded subspecialist had concordant interpretations for all test cases compared with the reference. In conclusion, our ML-based approach has similar performance for endoleak diagnosis relative to subspecialists and superior performance compared with generalists.

Recently developed processes have enabled bottom-up chemical synthesis of graphene nanoribbons (GNRs) with precise atomic structure. These GNRs are ideal candidates for electronic devices because of their uniformity, extremely narrow width below 1 nm, atomically perfect edge structure, and desirable electronic properties. Here, we demonstrate nano-scale chemically synthesized GNR field-effect transistors, made possible by development of a reliable layer transfer process. We observe strong environmental sensitivity and unique transport behavior characteristic of sub-1 nm width GNRs. © 2013 AIP Publishing LLC.

Deep-learning language models have shown promise in various biotechnological applications, including protein design and engineering. Here we describe ProGen, a language model that can generate protein sequences with a predictable function across large protein families, akin to generating grammatically and semantically correct natural language sentences on diverse topics. The model was trained on 280 million protein sequences from >19,000 families and is augmented with control tags specifying protein properties. ProGen can be further fine-tuned to curated sequences and tags to improve controllable generation performance of proteins from families with sufficient homologous samples. Artificial proteins fine-tuned to five distinct lysozyme families showed similar catalytic efficiencies as natural lysozymes, with sequence identity to natural proteins as low as 31.4%. ProGen is readily adapted to diverse protein families, as we demonstrate with chorismate mutase and malate dehydrogenase.