Exploring chemical reactivity through a combined conceptual DFT and ELF topology approach.

CONTEXT: In a proof-of-concept study, we explore how a combined approach using the topology of the electron localization function (ELF) and the condensed dual descriptor (DD) function can guide the optimal orientation between reactants and mimic the potential energy surfaces of molecular systems at the beginning of the chemical pathway. The DD has been chosen for its ability to evaluate the regioselectivity of neutral and soft species and to potentially mimic the interaction energy obtained from the mutual interactions between nucleophilic and electrophilic regions of the building blocks under perturbative theory. METHOD: Our method has been illustrated with examples in which the optimal orientation of several systems can be successfully identified. The limitations of the presented model in predicting chemical reactivity are outlined in particular the influence of the selected condensation scheme.

Study on the differentiation of sensory quality of mainstream Jiang-flavor baijiu in the Chinese market based on Pivot Profile.

Jiang-flavor baijiu (JFB) is a prominent type of Chinese baijiu, known for its unique flavor attributes, sensory experience, and high tasting value. Previous research has mainly focused on the detection and identification of its flavor substances, but in-depth studies on the precise sensory description and differentiation of its flavor qualities are still lacking. In this study, a rapid sensory analysis method, Pivot Profile (PP), was applied to 30 mainstream JFBs in the Chinese market, generating 91 sensory attributes with independent definitions, from which 29 main sensory attributes were established that were easy to perceive and descriptive, as well as convenient for transmitting their sensory qualities and distinguishing differences in price and production region, including color (one descriptor), aroma (21 descriptors), taste, and mouthfeel (seven descriptors). The nine key sensory attributes that distinguish JFB quality are as follows: Jiang, Grain, Chen, Qu, Rancid, Acid, Sweet, Fullness, and Harmony. It was found that price was positively correlated with sensory quality, with greater variation in the quality of samples within the medium price range (RMB 500-1000). All samples from MTCQ1 (the core production area of Maotai Town) performed better in sensory quality. In addition, salted vegetable showed a high degree of regional characteristics, concentrated in most of the production regions of Guizhou Province. Aroma attributes were more suitable than taste and mouthfeel as sensory indicators for distinguishing production regions. This study has opened the direction of systematic construction of sensory description of JFB and provided a successful case for the evaluation of Chinese baijiu using novel sensory analysis techniques.

Rational Design of Diatomic Active Sites for Elucidating Oxygen Evolution Reaction Performance Trends.

Diatomic catalysts, especially those with heteronuclear active sites, have recently attracted significant attention for their advantages over single-atom catalysts in reactions with relatively high energy barrier, e.g. oxygen evolution reaction. Rational design and synthesis of heteronuclear diatomic catalysts are of immense significance but have so far been plagued by the lack of a definitive correlation between structure and catalytic properties. Here, we report macrocyclic precursor constrained strategy to fabricate series of transition metal (MT, Ni, Co, Fe, Mn, or Cu)-noble (MN, Ir or Ru) centers in carbon material. One notable performance trend is observed in the order of Cu-MN < Mn-MN < Fe-MN < MN < Co-MN < Ni-MN. However, the pathway has been not altered, still following the traditional adsorption reaction mechanism. The effect of the MT atoms on the performances could possibly originate from the distinct adsorption/desorption behaviors of key intermediates (i.e. *OH, *O and/or *OOH), strongly implying that ΔG*OOH-ΔG*OH could be used as the performance descriptor. We believe that our work provides useful strategy for synthesis of diatomic active sites with sole coordination configuration and varied composition, and in-depth insight to their catalytic mechanism, which could be used for further optimization of diatomic catalysts towards oxygen electrocatalysis.

Virtual screening of drug materials for pharmaceutical tablet manufacturability with reference to sticking.

The manufacturing of pharmaceutical solid dosage forms, such as tablets involves a large number of successive processing operations including crystallisation of the drug substance, granulation, drying, milling, mixing of the formulation, and compaction. Each step is fraught with manufacturing problems. Undesired adhesion of powders to the surface of the compaction tooling, known as sticking, is a frequent and highly disruptive problem that occurs at the very end of the process chain when the tablet is formed. As an alternative to the mechanistic approaches to address sticking, we introduce two different machine learning strategies to predict sticking directly from the chemical formula of the drug substance, represented by molecular descriptors. An empirical database for sticking behaviour was developed and used to train the machine learning (ML) algorithms to predict sticking properties from molecular descriptors. The ML model has successfully classified sticking/non-sticking behaviour of powders with 100% separation. Predictions were made for materials in the handbook of Pharmaceutical Excipients and a subset of molecules included in the ChemBL database, demonstrating the potential use of machine learning approaches to screen for sticking propensity early at drug discovery and development stages. This is the first-time molecular descriptors and machine learning were used to predict and screen for sticking behaviour. The method has potential to transform the development of medicines by providing manufacturability information at drug screening stage and is potentially applicable to other manufacturing problems controlled by the chemistry of the drug substance.

Intrinsic Activity Identification of Noble Metal Single-Sites for Electrocatalytic Chlorine Evolution.

Single-atom catalysts with maximal atom-utilization have emerged as promising alternatives for chlorine evolution reaction (CER) toward valuable Cl2 production. However, understanding their intrinsic CER activity have so far been plagued due to the lack of well-defined atomic structure controlling. Herein, we prepare and identify a series of atomically dispersed noble metals (e.g., Pt, Ir, Ru) in nitrogen-doped nanocarbons (M1-N-C) with an identical M-N4 moiety, which allows objective activity evaluation. Electrochemical experiments, operando Raman spectroscopy, and quasi-in situ electron paramagnetic resonance spectroscopy analyses collectively reveal that all the three M1-N-C proceed the CER via a direct Cl-mediated Vomer-Heyrovský mechanism with reactivity following the trend of Pt1-N-C > Ir1-N-C > Ru1-N-C. Density functional theory (DFT) calculations reveal that this activity trend is governed by the binding strength of Cl*-Cl intermediate (ΔGCl*-Cl) on M-N4 sites (Pt < Ir < Ru) featuring distinct d-band centers, providing a reliable thermodynamic descriptor for rational design of single metal sites toward Cl2 electrosynthesis.

ProSol-multi: Protein solubility prediction via amino acids multi-level correlation and discriminative distribution.

Protein solubility prediction is useful for the careful selection of highly effective candidate proteins for drug development. In recombinant proteins synthesis, solubility prediction is valuable for optimizing key protein characteristics, including stability, functionality, and ease of purification. It contains valuable information about potential biomarkers or therapeutic targets and helps in early forecasting of neurodegenerative diseases, cancer, and cardiovascular disorders. Traditional wet-lab experimental protein solubility prediction approaches are error-prone, time-consuming, and costly. Researchers harnessed the competence of Artificial Intelligence approaches for replacing experimental approaches with computational predictors. These predictors inferred the solubility of proteins by analyzing amino acids distributions in raw protein sequences. There is still a lot of room for the development of robust computational predictors because existing predictors remain fail in extracting comprehensive discriminative distribution of amino acids. To more precisely discriminate soluble proteins from insoluble proteins, this paper presents ProSol-Multi predictor that makes use of a novel MLCDE encoder and Random Forest classifier. MLCDE encoder transforms protein sequences into informative statistical vectors by capturing amino acids multi-level correlation and discriminative distribution within raw protein sequences. The performance of proposed encoder is evaluated against 56 existing protein sequence encoding methods on a widely used protein solubility prediction benchmark dataset under two different experimental settings namely intrinsic and extrinsic. Intrinsic evaluation reveals that from all sequence encoders, proposed MLCDE encoder manages to generate non-overlapping clusters of soluble and insoluble classes. In extrinsic evaluation, 10 machine learning classifiers achieve better performance with proposed MLCDE encoder as compared to 56 existing protein sequence encoders. Moreover, across 4 public benchmark datasets, proposed ProSol-Multi predictor outshines 20 existing predictors by an average accuracy of 3%, MCC and AU-ROC of 2%. ProSol-Multi interactive web application is available at https://sds_genetic_analysis.opendfki.de/ProSol-Multi.

Metadata recommendations for light logging and dosimetry datasets.

Background: Light exposure significantly impacts human health, regulating our circadian clock, sleep-wake cycle and other physiological processes. With the emergence of wearable light loggers and dosimeters, research on real-world light exposure effects is growing. There is a critical need to standardize data collection and documentation across studies. Results: This article proposes a new metadata descriptor designed to capture crucial information within personalized light exposure datasets collected with wearable light loggers and dosimeters. The descriptor, developed collaboratively by international experts, has a modular structure for future expansion and customization. It covers four key domains: study design, participant characteristics, dataset details, and device specifications. Each domain includes specific metadata fields for comprehensive documentation. The user-friendly descriptor is available in JSON format. A web interface simplifies generating compliant JSON files for broad accessibility. Version control allows for future improvements. Conclusions: Our metadata descriptor empowers researchers to enhance the quality and value of their light dosimetry datasets by making them FAIR (findable, accessible, interoperable and reusable). Ultimately, its adoption will advance our understanding of how light exposure affects human physiology and behaviour in real-world settings.

Bridging the gap: Multi-omics profiling of brain tissue in Alzheimer's disease and older controls in multi-ethnic populations.

INTRODUCTION: Multi-omics studies in Alzheimer's disease (AD) revealed many potential disease pathways and therapeutic targets. Despite their promise of precision medicine, these studies lacked Black Americans (BA) and Latin Americans (LA), who are disproportionately affected by AD. METHODS: To bridge this gap, Accelerating Medicines Partnership in Alzheimer's Disease (AMP-AD) expanded brain multi-omics profiling to multi-ethnic donors. RESULTS: We generated multi-omics data and curated and harmonized phenotypic data from BA (n = 306), LA (n = 326), or BA and LA (n = 4) brain donors plus non-Hispanic White (n = 252) and other (n = 20) ethnic groups, to establish a foundational dataset enriched for BA and LA participants. This study describes the data available to the research community, including transcriptome from three brain regions, whole genome sequence, and proteome measures. DISCUSSION: The inclusion of traditionally underrepresented groups in multi-omics studies is essential to discovering the full spectrum of precision medicine targets that will be pertinent to all populations affected with AD. HIGHLIGHTS: Accelerating Medicines Partnership in Alzheimer's Disease Diversity Initiative led brain tissue profiling in multi-ethnic populations. Brain multi-omics data is generated from Black American, Latin American, and non-Hispanic White donors. RNA, whole genome sequencing and tandem mass tag proteomicsis completed and shared. Multiple brain regions including caudate, temporal and dorsolateral prefrontal cortex were profiled.

Electronic descriptor to identify the activity of SACs for e-NRR and effect of BF3 as electrolyte ion.

Electrochemical nitrogen reduction reaction (e-NRR) is an eco-friendly alternative approach to generate ammonia under ambient conditions, with very low power supply. But, developing of an efficient catalyst by suppressing parallel hydrogen evolution reaction as well as avoiding the catalysts poisoning either by hydrogen or electrolyte ion is an open question. So, in order to screen the single atom catalysts (SACs) for the e-NRR, we proposed a descriptor-based approach using density functional theory (DFT) based calculations. We investigated total 24 different SACs of types TM-Pc, TM-N3C1, TM-N2C2, TM-NC3 and TM-N4, considering transition metal (TM). We have considered mainly BF3 ion to understand the role of electrolyte and extended the study for four more electrolyte ions, Cl, ClO4, SO4, OH. Herein, to predict catalytic activity for a given catalyst we have tested 16 different electronic parameters. Out of those, electronic parameter dxz↓ occupancy, identified as electronic descriptor, is showing an excellent linear correlation with catalytic activity (R2 = 0.86). Furthermore, the selectivity of e-NRR over HER is defined by using an energy parameter ∆G*H-∆G*NNH. Further, the electronic descriptor (dxz↓ occupancy) can be used to predict promising catalysts for e-NRR, thus reducing the efforts on designing future single atom catalysts (SACs).

Neural network-based adaptive fault-tolerant control for nonlinear systems with uncertainties.

This paper proposes a novel fault-tolerant control (FTC) scheme for real-time uncertainty estimation in nonlinear systems. It addresses the challenges arising from nonlinear dynamics in system inputs, states, and outputs, along with measurement uncertainties, within an output feedback framework. Our approach leverages two key components: 1) A neural network NN descriptor-based observer: this novel observer concurrently estimates both system states and sensor uncertainties. It is particularly capable of handling unbounded sensor uncertainties in specific situations. It utilizes NNs as universal approximators to capture the system's complex nonlinearities. 2) A robust model reference tracking controller: this controller employs the estimated states from the NN descriptor-based observer to achieve the desired system performance despite the existence of uncertainties. It exhibits robustness, guaranteeing system stability and asymptotic state tracking to a given reference model. The efficacy of the proposed FTC scheme is validated through theoretical analysis and its application to two real-world case studies.

Development of an effective QSAR-based hazard threshold prediction model for the ecological risk assessment of aromatic hydrocarbon compounds.

The insufficient hazard thresholds of specific individual aromatic hydrocarbon compounds (AHCs) with diverse structures limit their ecological risk assessment. Thus, herein, quantitative structure-activity relationship (QSAR) models for estimating the hazard threshold of AHCs were developed based on the hazardous concentration for 5% of species (HC5) determined using the optimal species sensitivity distribution models and on the molecular descriptors calculated via the PADEL software and ORCA software. Results revealed that the optimal QSAR model, which involved eight descriptors, namely, Zagreb, GATS2m, VR3_Dzs, AATSC2s, GATS2c, ATSC2i, ω, and Vm, displayed excellent performance, as reflected by an optimal goodness of fit (R2adj = 0.918), robustness (Q2LOO = 0.869), and external prediction ability (Q2F1 = 0.760, Q2F2 = 0.782, and Q2F3 = 0.774). The hazard thresholds estimated using the optimal QSAR model were approximately close to the published water quality criteria developed by different countries and regions. The quantitative structure-toxicity relationship demonstrated that the molecular descriptors associated with electrophilicity and topological and electrotopological properties were important factors that affected the risks of AHCs. A new and reliable approach to estimate the hazard threshold of ecological risk assessment for various aromatic hydrocarbon pollutants was provided in this study, which can be widely popularised to similar contaminants with diverse structures.

Unveiling critical structural features for effective HDAC8 inhibition: a comprehensive study using quantitative read-across structure-activity relationship (q-RASAR) and pharmacophore modeling.

Histone deacetylases constitute a group of enzymes that participate in several biological processes. Notably, inhibiting HDAC8 has become a therapeutic strategy for various diseases. The current inhibitors for HDAC8 lack selectivity and target multiple HDACs. Consequently, there is a growing recognition of the need for selective HDAC8 inhibitors to enhance the effectiveness of therapeutic interventions. In our current study, we have utilized a multi-faceted approach, including Quantitative Structure-Activity Relationship (QSAR) combined with Quantitative Read-Across Structure-Activity Relationship (q-RASAR) modeling, pharmacophore mapping, molecular docking, and molecular dynamics (MD) simulations. The developed q-RASAR model has a high statistical significance and predictive ability (Q2F1:0.778, Q2F2:0.775). The contributions of important descriptors are discussed in detail to gain insight into the crucial structural features in HDAC8 inhibition. The best pharmacophore hypothesis exhibits a high regression coefficient (0.969) and a low root mean square deviation (0.944), highlighting the importance of correctly orienting hydrogen bond acceptor (HBA), ring aromatic (RA), and zinc-binding group (ZBG) features in designing potent HDAC8 inhibitors. To confirm the results of q-RASAR and pharmacophore mapping, molecular docking analysis of the five potent compounds (44, 54, 82, 102, and 118) was performed to gain further insights into these structural features crucial for interaction with the HDAC8 enzyme. Lastly, MD simulation studies of the most active compound (54, mapped correctly with the pharmacophore hypothesis) and the least active compound (34, mapped poorly with the pharmacophore hypothesis) were carried out to validate the observations of the studies above. This study not only refines our understanding of essential structural features for HDAC8 inhibition but also provides a robust framework for the rational design of novel selective HDAC8 inhibitors which may offer insights to medicinal chemists and researchers engaged in the development of HDAC8-targeted therapeutics.

FP-Zernike: An Open-source Structural Database Construction Toolkit for Fast Structure Retrieval.

The release of AlphaFold2 has sparked a rapid expansion in protein model databases. Efficient protein structure retrieval is crucial for the analysis of structure models, while measuring the similarity between structures is the key challenge in structural retrieval. Although existing structure alignment algorithms can address this challenge, they are often time-consuming. Currently, the state-of-the-art approach involves converting protein structures into three-dimensional (3D) Zernike descriptors and assessing similarity using Euclidean distance. However, the methods for computing 3D Zernike descriptors mainly rely on structural surfaces and are predominantly web-based, thus limiting their application in studying custom datasets. To overcome this limitation, we developed FP-Zernike, a user-friendly toolkit for computing different types of Zernike descriptors based on feature points. Users simply need to enter a single line of command to calculate the Zernike descriptors of all structures in customized datasets. FP-Zernike outperforms the leading method in terms of retrieval accuracy and binary classification accuracy across diverse benchmark datasets. In addition, we showed the application of FP-Zernike in the construction of the descriptor database and the protocol used for the Protein Data Bank (PDB) dataset to facilitate the local deployment of this tool for interested readers. Our demonstration contained 590,685 structures, and at this scale, our system required only 4-9 s to complete a retrieval. The experiments confirmed that it achieved the state-of-the-art accuracy level. FP-Zernike is an open-source toolkit, with the source code and related data accessible at https://ngdc.cncb.ac.cn/biocode/tools/BT007365/releases/0.1, as well as through a webserver at http://www.structbioinfo.cn/.

Design of Single-Atom Catalysts on C5N2 for Nitrogen Fixation at Ambient Conditions: A First-Principles Study.

Single atom catalysts (SACs) exhibit the flexible coordination structure of the active site and high utilization of active atoms, making them promising candidates for nitrogen reduction reaction (NRR) under ambient conditions. By the aid of first-principles calculations based on DFT, we have systematically explored the NRR catalytic behavior of thirteen 4d- and 5d-transition metal atoms anchored on 2D porous graphite carbon nitride C 5 ${_5 }$ N 2 ${_2 }$ . With high selectivity and outstanding activity, Zr, Nb, Mo, Ta, W and Re-doped C 5 ${_5 }$ N 2 ${_2 }$ are identified as potential nominees for NRR. Particularly, Mo@C 5 ${_5 }$ N 2 ${_2 }$ possesses an impressive low limiting potential of -0.39 V (corresponding to a very low temperature and atmospheric pressure), featuring the potential determining step involving *N-N transitions to *N-NH via the distal path. The catalytic performance of TM@C 5 ${_5 }$ N 2 ${_2 }$ can be well characterized by the adsorption strength of intermediate *N 2 ${_2 }$ H. Moreover, there exists a volcanic relationship between the catalytic property U L ${_{\rm{L}} }$ and the structure descriptor Ψ ${{{\Psi }}}$ , which validates the robustness and universality of Ψ ${{{\Psi }}}$ , combined with our previous study. This work sheds light on the design of SACs with eminent NRR performance.

Robust Point Cloud Registration for Aircraft Engine Pipeline Systems.

Aircraft engine systems are composed of numerous pipelines. It is crucial to regularly inspect these pipelines to detect any damages or failures that could potentially lead to serious accidents. The inspection process typically involves capturing complete 3D point clouds of the pipelines using 3D scanning techniques from multiple viewpoints. To obtain a complete and accurate representation of the aircraft pipeline system, it is necessary to register and align the individual point clouds acquired from different views. However, the structures of aircraft pipelines often appear similar from different viewpoints, and the scanning process is prone to occlusions, resulting in incomplete point cloud data. The occlusions pose a challenge for existing registration methods, as they can lead to missing or wrong correspondences. To this end, we present a novel registration framework specifically designed for aircraft pipeline scenes. The proposed framework consists of two main steps. First, we extract the point feature structure of the pipeline axis by leveraging the cylindrical characteristics observed between adjacent blocks. Then, we design a new 3D descriptor called PL-PPFs (Point Line-Point Pair Features), which combines information from both the pipeline features and the engine assembly line features within the aircraft pipeline point cloud. By incorporating these relevant features, our descriptor enables accurate identification of the structure of the engine's piping system. Experimental results demonstrate the effectiveness of our approach on aircraft engine pipeline point cloud data.

Place Recognition through Multiple LiDAR Scans Based on the Hidden Markov Model.

Autonomous driving systems for unmanned ground vehicles (UGV) operating in enclosed environments strongly rely on LiDAR localization with a prior map. Precise initial pose estimation is critical during system startup or when tracking is lost, ensuring safe UGV operation. Existing LiDAR-based place recognition methods often suffer from reduced accuracy due to only matching descriptors from individual LiDAR keyframes. This paper proposes a multi-frame descriptor-matching approach based on the hidden Markov model (HMM) to address this issue. This method enhances the place recognition accuracy and robustness by leveraging information from multiple frames. Experimental results from the KITTI dataset demonstrate that the proposed method significantly enhances the place recognition performance compared with the scan context-based single-frame descriptor-matching approach, with an average performance improvement of 5.8% and with a maximum improvement of 15.3%.

Subtraction Descriptors in Machine Learning for Optimizing the Cocatalyst Effect of Cobalt Phosphate on Hematite Photoanodes.

In the quest for sustainable energy solutions, the optimization of the photoelectrochemical (PEC) performance of hematite photoanodes through cocatalysts represents a promising avenue. This study introduces a novel machine learning approach, leveraging subtraction descriptors, to isolate and quantify the specific effects of cobalt phosphate (Co-Pi) as a cocatalyst on hematite's PEC performance. By integrating data from various analytical techniques, including photoelectrochemical impedance spectroscopy and ultraviolet-visible spectroscopy, with advanced machine learning models, we successfully predicted the PEC performance enhancement attributed to Co-Pi. The Gaussian process regression (GPR) model emerged as the most effective, revealing the critical influence of the interfacial resistance, bulk resistance, and interfacial capacitance on the PEC performance. These findings underscore the potential of cocatalysts in improving charge separation and extending charge carrier lifetimes, thereby boosting the efficiency of photocatalytic reactions. This study not only advances our understanding of the cocatalyst effect in photocatalytic systems but also demonstrates the power of machine learning in modifying complex materials and guiding the development of optimized photocatalytic materials. The implications of this research extend beyond hematite photoanodes, offering a generalizable framework for enhancing the photoelectrochemical properties of a wide range of material modifications such as cocatalyst deposition, doping, and passivation.

NanoConstruct: A web application builder of ellipsoidal nanoparticles for the investigation of their crystal growth, stability, and the calculation of atomistic descriptors.

NanoConstruct is a state-of-the-art computational tool that enables a) the digital construction of ellipsoidal neutral energy minimized nanoparticles (NPs) in vacuum through its graphical user-friendly interface, and b) the calculation of NPs atomistic descriptors. It allows the user to select NP's shape and size by inserting its ellipsoidal axes and rotation angle while the NP material is selected by uploading its Crystallography Information File (CIF). To investigate the stability of materials not yet synthesised, NanoConstruct allows the substitution of the chemical elements of an already synthesized material with chemical elements that belong into the same group and neighbouring rows of the periodic table. The process is divided into three stages: 1) digital construction of the unit cell, 2) digital construction of NP using geometry rules and keeping its stoichiometry and 3) energy minimization of the geometrically constructed NP and calculation of its atomistic descriptors. In this study, NanoConstruct was applied for the investigation of the crystal growth of Zirconia (ZrO2) NPs when in the rutile form. The most stable configuration and the crystal growth route were identified, showing a preferential direction for the crystal growth of ZrO2 in its rutile form. NanoConstruct is freely available through the Enalos Cloud Platform (https://enaloscloud.novamechanics.com/riskgone/nanoconstruct/).

A Machine Learning Approach for Predicting Caco-2 Cell Permeability in Natural Products from the Biodiversity in Peru.

BACKGROUND: Peru is one of the most biodiverse countries in the world, which is reflected in its wealth of knowledge about medicinal plants. However, there is a lack of information regarding intestinal absorption and the permeability of natural products. The human colon adenocarcinoma cell line (Caco-2) is an in vitro assay used to measure apparent permeability. This study aims to develop a quantitative structure-property relationship (QSPR) model using machine learning algorithms to predict the apparent permeability of the Caco-2 cell in natural products from Peru. METHODS: A dataset of 1817 compounds, including experimental log Papp values and molecular descriptors, was utilized. Six QSPR models were constructed: a multiple linear regression (MLR) model, a partial least squares regression (PLS) model, a support vector machine regression (SVM) model, a random forest (RF) model, a gradient boosting machine (GBM) model, and an SVM-RF-GBM model. RESULTS: An evaluation of the testing set revealed that the MLR and PLS models exhibited an RMSE = 0.47 and R2 = 0.63. In contrast, the SVM, RF, and GBM models showcased an RMSE = 0.39-0.40 and R2 = 0.73-0.74. Notably, the SVM-RF-GBM model demonstrated superior performance, with an RMSE = 0.38 and R2 = 0.76. The model predicted log Papp values for 502 natural products falling within the applicability domain, with 68.9% (n = 346) showing high permeability, suggesting the potential for intestinal absorption. Additionally, we categorized the natural products into six metabolic pathways and assessed their drug-likeness. CONCLUSIONS: Our results provide insights into the potential intestinal absorption of natural products in Peru, thus facilitating drug development and pharmaceutical discovery efforts.

Phenotyping grapevine resistance to downy mildew: deep learning as a promising tool to assess sporulation and necrosis.

BACKGROUND: Downy mildew is a plant disease that affects all cultivated European grapevine varieties. The disease is caused by the oomycete Plasmopara viticola. The current strategy to control this threat relies on repeated applications of fungicides. The most eco-friendly and sustainable alternative solution would be to use bred-resistant varieties. During breeding programs, some wild Vitis species have been used as resistance sources to introduce resistance loci in Vitis vinifera varieties. To ensure the durability of resistance, resistant varieties are built on combinations of these loci, some of which are unfortunately already overcome by virulent pathogen strains. The development of a high-throughput machine learning phenotyping method is now essential for identifying new resistance loci. RESULTS: Images of grapevine leaf discs infected with P. viticola were annotated with OIV 452-1 values, a standard scale, traditionally used by experts to assess resistance visually. This descriptor takes two variables into account the complete phenotype of the symptom: sporulation and necrosis. This annotated dataset was used to train neural networks. Various encoders were used to incorporate prior knowledge of the scale's ordinality. The best results were obtained with the Swin transformer encoder which achieved an accuracy of 81.7%. Finally, from a biological point of view, the model described the studied trait and identified differences between genotypes in agreement with human observers, with an accuracy of 97% but at a high-throughput 650% faster than that of humans. CONCLUSION: This work provides a fast, full pipeline for image processing, including machine learning, to describe the symptoms of grapevine leaf discs infected with P. viticola using the OIV 452-1, a two-symptom standard scale that considers sporulation and necrosis. If symptoms are frequently assessed by visual observation, which is time-consuming, low-throughput, tedious, and expert dependent, the method developed sweeps away all these constraints. This method could be extended to other pathosystems studied on leaf discs where disease symptoms are scored with ordinal scales.