With advancements in machine learning a large data scale, high throughput virtual screening has become a more attractive method for screening drug candidates. This study compared the accuracy of molecular descriptors from two cheminformatics Mordred and PaDEL, software libraries, in characterizing the chemo-structural composition of 53 compounds from the non-nucleoside reverse transcriptase inhibitors (NNRTI) class. The classification model built with the filtered set of descriptors from Mordred was superior to the model using PaDEL descriptors. This approach can accelerate the identification of hit compounds and improve the efficiency of the drug discovery pipeline.

The authors investigate the ability of machine learning models to developing new drug-like molecules by learning desired chemical properties versus simply generating molecules that similar to those in the training set.

These authors mathematically deduce a model that explains the interesting (and unintuitive) physical phenomenon that occurs when water falls.

Semantic segmentation - labelling each pixel in an image to a specific class- models require large amounts of manually labeled and collected data to train.

Since school bathrooms are widely suspected to be unsanitary, we wanted to compare the total amount of bacteria with the amount of bacteria that had ampicillin or streptomycin resistance across different school bathrooms in the Boston area. We hypothesized that because people interact with the faucet, outdoor handle, and indoor handle of the bathroom, based on whether or not they have washed their hands, there would be differences in the quantity of the bacteria presented on these surfaces. Therefore, we predicted certain surfaces of the bathroom would be less sanitary than others.

Here, the authors explored how the sale and use of electric vehicles could reduce emissions from the transport industry in Canada. By fitting the sale of total of electric vehicles with an exponential model, the authors predicted the number of electric vehicle sales through 2030 and related that to the average emission for such vehicles. Ultimately, they found that the sale and use of electric vehicles alone would likely not meet the 45% reduction in emissions from the transport industry suggested by the Canadian government

Climate change is predicted to increase the frequency of severe thunderstorm events in coming years. In this study, the authors hypothesized that (i) the majority of severe thunderstorm events will occur in the summer months in all states examined for all years analyzed, (ii) climate change will cause an unusual number of severe thunderstorm events in winter months in all states, (iii) thundersnow would be observed in Colorado, and (iv.) there would be no difference in the number of severe thunderstorm events between states in any given year examined. They classified lightning seasons in all states observed, with the most severe thunderstorm events occurring in May, June, July, and August. Colorado, New Jersey, Washington, and West Virginia were found to have severe thunderstorm events in the winter, which could be explained by increased winter storms due to climate change (1). Overall, they highlight the importance of quantifying when lightning seasons occur to avoid lightning-related injuries or death.

Glioblastoma Multiforme (GBM) is the most malignant brain tumor with the highest fraction of genome alterations (FGA), manifesting poor disease-free status (DFS) and overall survival (OS). We explored The Cancer Genome Atlas (TCGA) and cBioportal public dataset- Firehose legacy GBM to study DNA repair genes Activating Signal Cointegrator 1 Complex Subunit 3 (ASCC3) and Alpha-Ketoglutarate-Dependent Dioxygenase AlkB Homolog 3 (ALKBH3). To test our hypothesis that these genes have correlations with FGA and can better determine prognosis and survival, we sorted the dataset to arrive at 254 patients. Analyzing using RStudio, both ASCC3 and ALKBH3 demonstrated hypomethylation in 82.3% and 61.8% of patients, respectively. Interestingly, low mRNA expression was observed in both these genes. We further conducted correlation tests between both methylation and mRNA expression of these genes with FGA. ASCC3 was found to be negatively correlated, while ALKBH3 was found to be positively correlated, potentially indicating contrasting dysregulation of these two genes. Prognostic analysis showed the following: ASCC3 hypomethylation is significant with DFS and high ASCC3 mRNA expression to be significant with OS, demonstrating ASCC3’s potential as disease prediction marker.

Quorum sensing (QS) is the process in which bacteria recognize and respond to the surrounding cell density, and it can be inhibited by certain antimicrobial substances. This study showed that illumination intensity data is insufficient for evaluating QS activity without proper statistical modeling. It concluded that modeling illumination intensity through time provides a more accurate evaluation of QS activity than conventional cross-sectional analysis.

Osteosarcoma is a type of bone cancer that affects young adults and children. Early diagnosis of osteosarcoma is crucial to successful treatment. The current methods of diagnosis, which include imaging tests and biopsy, are time consuming and prone to human error. Hence, we used deep learning to extract patterns and detect osteosarcoma from histological images. We hypothesized that the combination of two different technologies (transfer learning and data augmentation) would improve the efficacy of osteosarcoma detection in histological images. The dataset used for the study consisted of histological images for osteosarcoma and was quite imbalanced as it contained very few images with tumors. Since transfer learning uses existing knowledge for the purpose of classification and detection, we hypothesized it would be proficient on such an imbalanced dataset. To further improve our learning, we used data augmentation to include variations in the dataset. We further evaluated the efficacy of different convolutional neural network models on this task. We obtained an accuracy of 91.18% using the transfer learning model MobileNetV2 as the base model with various geometric transformations, outperforming the state-of-the-art convolutional neural network based approach.