Optimizing feature selection with gradient boosting machines in PLS regression for predicting moisture and protein in multi-country corn kernels via NIR spectroscopy.

Differences in moisture and protein content impact both nutritional value and processing efficiency of corn kernels. Near-infrared (NIR) spectroscopy can be used to estimate kernel composition, but models trained on a few environments may underestimate error rates and bias. We assembled corn samples from diverse international environments and used NIR with chemometrics and partial least squares regression (PLSR) to determine moisture and protein. The potential of five feature selection methods to improve prediction accuracy was assessed by extracting sensitive wavelengths. Gradient boosting machines (GBMs), particularly CatBoost and LightGBM, were found to effectively select crucial wavelengths for moisture (1409, 1900, 1908, 1932, 1953, 2174 nm) and protein (887, 1212, 1705, 1891, 2097, 2456 nm). SHAP plots highlighted significant wavelength contributions to model prediction. These results illustrate GBMs' effectiveness in feature engineering for agricultural and food sector applications, including developing multi-country global calibration models for moisture and protein in corn kernels.

Imipramine and olanzapine block apoE4-catalyzed polymerization of Aß and show evidence of improving Alzheimer's disease cognition.

BACKGROUND: The apolipoprotein E (APOE) ε4 allele confers the strongest risk for late-onset Alzheimer's disease (AD) besides age itself, but the mechanisms underlying this risk are debated. One hypothesis supported by evidence from multiple labs is that apoE4 binds to the amyloid-ß (Aß) peptide and catalyzes its polymerization into neurotoxic oligomers and fibrils. Inhibiting this early step in the amyloid cascade may thereby reduce or prevent neurodegeneration and AD. METHODS: Using a design of experiments (DOE) approach, we developed a high-throughput assay to identify inhibitors of apoE4-catalyzed polymerization of Aß into oligomers and fibrils. We used it to screen the NIH Clinical Collection of small molecule drugs tested previously in human clinical trials. We then evaluated the efficacy and cytotoxicity of the hit compounds in primary neuron models of apoE4-induced Aß and phosphorylated tau aggregation. Finally, we performed retrospective analyses of the National Alzheimer's Coordinating Center (NACC) clinical dataset, using Cox regression and Cox proportional hazards models to determine if the use of two FDA-approved hit compounds was associated with better cognitive scores (Mini-Mental State Exam), or improved AD clinical diagnosis, when compared with other medications of the same clinical indication. RESULTS: Our high-throughput screen identified eight blood-brain barrier (BBB)-permeable hit compounds that reduced apoE4-catalyzed Aß oligomer and fibril formation in a dose-dependent manner. Five hit compounds were non-toxic toward cultured neurons and also reduced apoE4-promoted Aß and tau neuropathology in a dose-dependent manner. Three of the five compounds were determined to be specific inhibitors of apoE4, whereas the other two compounds were Aß or tau aggregation inhibitors. When prescribed to AD patients for their normal clinical indications, two of the apoE4 inhibitors, imipramine and olanzapine, but not other (non-hit) antipsychotic or antidepressant medications, were associated with improvements in cognition and clinical diagnosis, especially among APOE4 carriers. CONCLUSIONS: The critical test of any proposed AD mechanism is whether it leads to effective treatments. Our high-throughput screen identified two promising FDA-approved drugs, imipramine and olanzapine, which have no structural, functional, or clinical similarities other than their shared ability to inhibit apoE4-catalyzed Aß polymerization, thus identifying this mechanism as an essential contribution of apoE4 to AD.

Deep learning strategies for addressing issues with small datasets in 2D materials research: Microbial Corrosion.

Protective coatings based on two dimensional materials such as graphene have gained traction for diverse applications. Their impermeability, inertness, excellent bonding with metals, and amenability to functionalization renders them as promising coatings for both abiotic and microbiologically influenced corrosion (MIC). Owing to the success of graphene coatings, the whole family of 2D materials, including hexagonal boron nitride and molybdenum disulphide are being screened to obtain other promising coatings. AI-based data-driven models can accelerate virtual screening of 2D coatings with desirable physical and chemical properties. However, lack of large experimental datasets renders training of classifiers difficult and often results in over-fitting. Generate large datasets for MIC resistance of 2D coatings is both complex and laborious. Deep learning data augmentation methods can alleviate this issue by generating synthetic electrochemical data that resembles the training data classes. Here, we investigated two different deep generative models, namely variation autoencoder (VAE) and generative adversarial network (GAN) for generating synthetic data for expanding small experimental datasets. Our model experimental system included few layered graphene over copper surfaces. The synthetic data generated using GAN displayed a greater neural network system performance (83-85% accuracy) than VAE generated synthetic data (78-80% accuracy). However, VAE data performed better (90% accuracy) than GAN data (84%-85% accuracy) when using XGBoost. Finally, we show that synthetic data based on VAE and GAN models can drive machine learning models for developing MIC resistant 2D coatings.