Plant diseases can cause up to 50% crop yield loss for the popular tomato plant. A mobile device-based method to identify diseases from photos of symptomatic leaves via computer vision can be more effective due to its convenience and accessibility. To enable a practical mobile solution, a “shallow” convolutional neural networks (CNNs) with few layers, and thus low computational requirement but with high accuracy similar to the deep CNNs is needed. In this work, we explored if such a model was possible.

Molecular dynamics (MD) simulations are a great tool to model and study complex biological systems. In this paper, the authors use MD simulations to construct and simulate a model of the periplasmic space, the peptidoglycan layer and its associated proteins, in an Escherichia coli cell.

With molecular energy being an integral element to the study of molecules and molecular interactions, computational methods to determine molecular energy are used for the preservation of time and resources. However, these computational methods have high demand for computer resources, limiting their widespread feasibility. The authors of this study employed machine learning to address this disadvantage, utilizing neural networks trained on different representations of molecules to predict molecular properties without the requirement of computationally-intensive processing. In their findings, the authors determined the Feedforward Neural Network, trained by two separate models, as capable of predicting molecular energy with limited prediction error.

Here, based on the identification of androgapholide as a potential therapeutic treatment against cancer, Alzheimer's disease, diabetes, and multiple sclerosis, due to its ability to inhibit a signaling pathway in immune system function, the authors sought ways to optimize the natural product human systems by manipulating its chemical structure. Through the semisynthesis of a natural product along with computational studies, the authors developed an understanding of the kinetic mechanisms of andrographolide and semisynthetic analogs in the context of Michael additions.

Natural selection shapes the evolution of all organisms, and one question of interest is whether natural selection will reach a "stopping point": a stable, ideal, value for any particular trait. Madhan and Kanagavel tackle this question by building a computer simulation of trait evolution in organisms.

Cell-free biologicals are a novel method of treating clinical conditions which involve chronic inflammation such as tendonitis and osteoarthritis. This study compared platelet-derived growth factor (PDGF) and vascular endothelial growth factor (VEGF) in platelet-rich plasma (PRP), activated PRP (aPRP), and platelet lysate (PL). It was hypothesized that PL would contain higher concentrations of growth factors than PRP and that different storage temperatures for PL would diminish cytokine expression. Results demonstrated PL had the highest concentrations of both cytokines, with concentrations slightly diminishing at-80C. aPRP and PRP demonstrated lower concentrations of PDGF and VEGF than PL.

It's time-consuming to complete the calculations that are used to study nuclear reactions and energy. To uncover which computational chemistry tools are useful for this challenge, Pan, Vaiyakarnam, Li, and McMahan investigated whether the Python-based Simulations of Chemistry Framework’s Hartree-Fock (PySCF) method is an efficient and accurate way to assess alkane molecules.

Modern artificial neural networks have been remarkably successful in various applications, from speech recognition to computer vision. However, it remains less clear whether they can implement abstract concepts, which are essential to generalization and understanding. To address this problem, the authors investigated the above vs. below task, a simple concept-based task that honeybees can solve, using a conventional neural network. They found that networks achieved 100% test accuracy when a visual target was presented below a black bar, however only 50% test accuracy when a visual target was presented below a reference shape.

Here, beginning from an initial interest in the possibility to use a computer to automatically solve a geometry diagram parser, the authors developed their own Fast Geometry Diagram Parser (FastGDP) that uses clustering and corner information. They compared their own methods to a more widely available, method, GeoSolver, finding their own to be an order of magnitude faster in most cases that they considered.

In this study, the authors determined whether tautomerization dynamics in protic and aprotic solvents displayed differences in reaction rates and in the proportion of the keto and enol tautomers present.