Water scarcity affects upwards of a billion people worldwide today. This project leverages the potential of capturing humidity to build a high-efficiency water condensation device that can generate water and be used for personal and commercial purposes. This compact environment-friendly device would have low power requirements, which would potentially allow it to utilize renewable energy sources and collect water at the most needed location.

Here based on an interest in fractals, the authors used a Julia Set Generator to consider a specific point on the Mandelbrot set with an associated coordinate. In this manner, they found that the complexity of the Mandelbrot and Julia Sets are governed by relatively simple rules, revealing that the intricate patterns of fractals can be defined by defined by simple rules and patterns.

Neural machine translation (NMT) is a software that uses neural network techniques to translate text from one language to another. However, one of the most famous NMT models—Google Translate—failed to give an accurate English translation of a famous Korean nursery rhyme, "Airplane" (비행기). The authors fine-tuned a pre-trained model first with a dataset from the lyrics domain, and then with a smaller dataset containing the rhythmical properties, to teach the model to translate rhythmically accurate lyrics. This stacked fine-tuning method resulted in an NMT model that could maintain the rhythmical characteristics of lyrics during translation while single fine-tuned models failed to do so.

Muons, one of the fundamental elementary particles, originate from the collision of cosmic rays with atmospheric particles and are also generated in particle accelerator collisions. In this study, Samson et al analyze the factors that influence muon flux and lifetime using Cosmic Ray Muon Detectors (CRMDs). Overall, the study suggests that water can be used to decrease muon flux and that scintillator orientation is a potential determinant of the volume of data collected in muon decay studies.

Free radical chain reactions result when atoms containing unpaired electrons bind with biomolecules and alter their biological functions, contributing to the progression of diseases such as atherosclerosis, cancer, and diabetes. Antioxidants, such as vitamin E and sulforaphane, are effective neutralizers of free radicals and prevent cellular damage. This present study is conducted to determine the relative effectiveness of sulforaphane against free radicals generated by hydrogen peroxide (H2O2) compared with the known antioxidant vitamin E.

Smoking generates free radicals and reactive oxygen species which induce cell damage and lipid peroxidation. This is linked to the development of oral cancer in chronic smokers. The authors of this study developed Quitpuff, simple colorimetric test to measure the extent of lipid peroxidation in saliva samples. This test detected salivary lipid peroxidation with 96% accuracy in test subjects and could serve as an inexpensive, non-invasive test for smokers to measure degree of salivary lipid peroxidation and potential risk of oral cancer.

Wang and Gong developed a novel dynamic gene-searching algorithm called Dynamic Gene Search (DyGS) to create a gene panel for each of the 12 cancers with the highest annual incidence and death rate. The 12 gene panels the DyGS algorithm selected used only 3.5% of the original gene mutation pool, while covering every patient sample. About 40% of each gene panel is druggable, which indicates that the DyGS-generated gene panels can be used for early cancer detection as well as therapeutic targets in treatment methods.

To address whether odor sensory circuits are organized topographically, the authors investigate whether the neuronal responses to similar odors amongst different mice mapped similarly in brain.

In this study, the authors tested the ability and accuracy of a neural net to identify patterns in complex number matrices.

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.