High-throughput Prediction Research Articles

High-throughput prediction of oral acute toxicity in Rat and Mouse of over 100,000 polychlorinated persistent organic pollutants (PC-POPs) by interpretable data fusion-driven machine learning global models

This study utilized available oral acute toxicity data in Rat and Mouse for polychlorinated persistent organic pollutants (PC-POPs) to construct data fusion-driven machine learning (ML) global models. Based on atom-centered fragments (ACFs), the collected high-throughput data overcame the applicability limitations, enabling accurate toxicity prediction for a wide range of PC-POPs series compounds using only single models. The data variances in the Rat training and test sets were 1.52 and 1.34, respectively, while for the Mouse, the values were 1.48 and 1.36, respectively. Genetic algorithm (GA) was used to build multiple linear regression (MLR) models and pre-screen descriptors, addressing the “black-box” problem prevalent in ML and enhancing model interpretability. The best ML models for Rat and Mouse achieved approximately 90 % prediction reliability for over 100,000 true untested compounds. Ultimately, a warning list of highly toxic compounds for eight categories of polychlorinated atom-centered fragments (PCACFs) was generated based on the prediction results. The analysis of descriptors revealed that dioxin analogs generally exhibited higher toxicity, because the heteroatoms and ring systems increased structural complexity and formed larger conjugated systems, contributing to greater oral acute toxicity. The present study provides valuable insights for guiding the subsequent in vivo tests, environmental risk assessment and the improvement of global governance system of pollutants.

HMPA: a pioneering framework for the noncanonical peptidome from discovery to functional insights.

Advancements in peptidomics have revealed numerous small open reading frames with coding potential and revealed that some of these micropeptides are closely related to human cancer. However, the systematic analysis and integration from sequence to structure and function remains largely undeveloped. Here, as a solution, we built a workflow for the collection and analysis of proteomic data, transcriptomic data, and clinical outcomes for cancer-associated micropeptides using publicly available datasets from large cohorts. We initially identified 19586 novel micropeptides by reanalyzing proteomic profile data from 3753 samples across 8 cancer types. Further quantitative analysis of these micropeptides, along with associated clinical data, identified 3065 that were dysregulated in cancer, with 370 of them showing a strong association with prognosis. Moreover, we employed a deep learning framework to construct a micropeptide-protein interaction network for further bioinformatics analysis, revealing that micropeptides are involved in multiple biological processes as bioactive molecules. Taken together, our atlas provides a benchmark for high-throughput prediction and functional exploration of micropeptides, providing new insights into their biological mechanisms in cancer. The HMPA is freely available at http://hmpa.zju.edu.cn.

High-throughput prediction of stalk cellulose and hemicellulose content in maize using machine learning and Fourier transform infrared spectroscopy

Cellulose and hemicellulose are key cross-linked carbohydrates affecting bioethanol production in maize stalks. Traditional wet chemical methods for their detection are labor-intensive, highlighting the need for high-throughput techniques. This study used Fourier transform infrared (FTIR) spectroscopy combined with machine learning (ML) algorithms on 200 large-scale maize germplasms to develop robust predictive models for stalk cellulose, hemicellulose and holocellulose content. We identified several peak height features correlated with three contents, used them as input data for model building. Four ML algorithms demonstrated higher predictive accuracy, achieving coefficient of determination (R2) ranging from 0.83 to 0.97. Notably, the Categorical Boosting algorithm yielded optimal models with coefficient of determination (R2) exceeding 0.91 for the training set and over 0.81 for the test set. The approach combined FTIR spectroscopy with ML algorithms offers a precise and high-throughput tool for predicting stalk cellulose, hemicellulose and holocellulose contents, benefiting maize genetic breeding for bioenergy and biofuels.

PPI-hotspotID for detecting protein–protein interaction hot spots from the free protein structure

Experimental detection of residues critical for protein–protein interactions (PPI) is a time-consuming, costly, and labor-intensive process. Hence, high-throughput PPI-hot spot prediction methods have been developed, but they have been validated using relatively small datasets, which may compromise their predictive reliability. Here, we introduce PPI-hotspotID, a novel method for identifying PPI-hot spots using the free protein structure, and validated it on the largest collection of experimentally confirmed PPI-hot spots to date. We explored the possibility of detecting PPI-hot spots using (i) FTMap in the PPI mode, which identifies hot spots on protein–protein interfaces from the free protein structure, and (ii) the interface residues predicted by AlphaFold-Multimer. PPI-hotspotID yielded better performance than FTMap and SPOTONE, a webserver for predicting PPI-hot spots given the protein sequence. When combined with the AlphaFold-Multimer-predicted interface residues, PPI-hotspotID yielded better performance than either method alone. Furthermore, we experimentally verified several PPI-hotspotID-predicted PPI-hot spots of eukaryotic elongation factor 2. Notably, PPI-hotspotID can reveal PPI-hot spots not obvious from complex structures, including those in indirect contact with binding partners. PPI-hotspotID serves as a valuable tool for understanding PPI mechanisms and aiding drug design. It is available as a web server (https://ppihotspotid.limlab.dnsalias.org/) and open-source code (https://github.com/wrigjz/ppihotspotid/).

PPI-hotspotID for detecting protein-protein interaction hot spots from the free protein structure.

Experimental detection of residues critical for protein-protein interactions (PPI) is a time-consuming, costly, and labor-intensive process. Hence, high-throughput PPI-hot spot prediction methods have been developed, but they have been validated using relatively small datasets, which may compromise their predictive reliability. Here, we introduce PPI-hotspotID, a novel method for identifying PPI-hot spots using the free protein structure, and validated it on the largest collection of experimentally confirmed PPI-hot spots to date. We explored the possibility of detecting PPI-hot spots using (i) FTMap in the PPI mode, which identifies hot spots on protein-protein interfaces from the free protein structure, and (ii) the interface residues predicted by AlphaFold-Multimer. PPI-hotspotID yielded better performance than FTMap and SPOTONE, a webserver for predicting PPI-hot spots given the protein sequence. When combined with the AlphaFold-Multimer-predicted interface residues, PPI-hotspotID yielded better performance than either method alone. Furthermore, we experimentally verified several PPI-hotspotID-predicted PPI-hot spots of eukaryotic elongation factor 2. Notably, PPI-hotspotID can reveal PPI-hot spots not obvious from complex structures, including those in indirect contact with binding partners. PPI-hotspotID serves as a valuable tool for understanding PPI mechanisms and aiding drug design. It is available as a web server (https://ppihotspotid.limlab.dnsalias.org/) and open-source code (https://github.com/wrigjz/ppihotspotid/).

High-Throughput Screening and Prediction of Nucleophilicity of Amines Using Machine Learning and DFT Calculations.

Nucleophilic index (NNu) as a significant parameter plays a crucial role in screening of amine catalysts. Indeed, the quantity and variety of amines are extensive. However, only limited amines exhibit an NNu value exceeding 4.0 eV, rendering them potential nucleophiles in chemical reactions. To address this issue, we proposed a computational method to quickly identify amines with high NNu values by using Machine Learning (ML) and high-throughput Density Functional Theory (DFT) calculations. Our approach commenced by training ML models and the exploration of Molecular Fingerprint methods as well as the development of quantitative structure-activity relationship (QSAR) models for the well-known amines based on NNu values derived from DFT calculations. Utilizing explainable Shapley Additive Explanation plots, we were able to determine the five critical substructures that significantly impact the NNu values of amine. The aforementioned conclusion can be applied to produce and cultivate 4920 novel hypothetical amines with high NNu values. The QSAR models were employed to predict the NNu values of 259 well-known and 4920 hypothetical amines, resulting in the identification of five novel hypothetical amines with exceptional NNu values (>4.55 eV). The enhanced NNu values of these novel amines were validated by DFT calculations. One novel hypothetical amine, H1, exhibits an unprecedentedly high NNu value of 5.36 eV, surpassing the maximum value (5.35 eV) observed in well-established amines. Our research strategy efficiently accelerates the discovery of the high nucleophilicity of amines using ML predictions, as well as the DFT calculations.

Predicting protein conformational motions using energetic frustration analysis and AlphaFold2

Proteins perform their biological functions through motion. Although high throughput prediction of the three-dimensional static structures of proteins has proved feasible using deep-learning-based methods, predicting the conformational motions remains a challenge. Purely data-driven machine learning methods encounter difficulty for addressing such motions because available laboratory data on conformational motions are still limited. In this work, we develop a method for generating protein allosteric motions by integrating physical energy landscape information into deep-learning-based methods. We show that local energetic frustration, which represents a quantification of the local features of the energy landscape governing protein allosteric dynamics, can be utilized to empower AlphaFold2 (AF2) to predict protein conformational motions. Starting from ground state static structures, this integrative method generates alternative structures as well as pathways of protein conformational motions, using a progressive enhancement of the energetic frustration features in the input multiple sequence alignment sequences. For a model protein adenylate kinase, we show that the generated conformational motions are consistent with available experimental and molecular dynamics simulation data. Applying the method to another two proteins KaiB and ribose-binding protein, which involve large-amplitude conformational changes, can also successfully generate the alternative conformations. We also show how to extract overall features of the AF2 energy landscape topography, which has been considered by many to be black box. Incorporating physical knowledge into deep-learning-based structure prediction algorithms provides a useful strategy to address the challenges of dynamic structure prediction of allosteric proteins.

PyAMPA: a high-throughput prediction and optimization tool for antimicrobial peptides.

This paper introduces PyAMPA, a new bioinformatics platform designed for the discovery and optimization of antimicrobial peptides (AMPs). It addresses the urgent need for new antimicrobials due to the rise of antibiotic-resistant infections. PyAMPA, with its five predictive modules -AMPScreen, AMPValidate, AMPSolve, AMPMutate and AMPOptimize, enables high-throughput screening of proteomes to identify potential AMP motifs and optimize them for clinical use. Its unique approach, combining prediction, design, and optimization tools, makes PyAMPA a robust solution for developing new AMP-based therapies, offering a significant advance in combatting antibiotic resistance.

Deep DNAshape webserver: prediction and real-time visualization of DNA shape considering extended k-mers.

Sequence-dependent DNA shape plays an important role in understanding protein-DNA binding mechanisms. High-throughput prediction of DNA shape features has become a valuable tool in the field of protein-DNA recognition, transcription factor-DNA binding specificity, and gene regulation. However, our widely used webserver, DNAshape, relies on statistically summarized pentamer query tables to query DNA shape features. These query tables do not consider flanking regions longer than two base pairs, and acquiring a query table for hexamers or higher-order k-mers is currently still unrealistic due to limitations in achieving sufficient statistical coverage in molecular simulations or structural biology experiments. A recent deep-learning method, Deep DNAshape, can predict DNA shape features at the core of a DNA fragment considering flanking regions of up to seven base pairs, trained on limited simulation data. However, Deep DNAshape is rather complicated to install, and it must run locally compared to the pentamer-based DNAshape webserver, creating a barrier for users. Here, we present the Deep DNAshape webserver, which has the benefits of both methods while being accurate, fast, and accessible to all users. Additional improvements of the webserver include the detection of user input in real time, the ability of interactive visualization tools and different modes of analyses. URL: https://deepdnashape.usc.edu.

Microbial community changes and metabolic pathways analysis during waste activated sludge and meat processing waste anaerobic co-digestion

Waste activated sludge (WAS) and meat processing waste (MPW) were acted as co-substrates in anaerobic co-digestion (AcD), and biochemical methane potential (BMP) test was carried out to investigate the methane production performances. Microbial community structure and metabolic pathways analyses were conducted by 16S rRNA high-throughput sequencing and functional prediction analysis. BMP test results indicated that AcD of 70% WAS+30% MPW and 50% WAS+50% MPW (VS/VS) could significantly improve methane yield to 371.05 mL/g VS and 599.61 mL/g VS, respectively, compared with WAS acting as sole substrate (191.87 mL/g VS). The results of microbial community analysis showed that Syntrophomonas and Petrimonas became the dominant bacteria genera, and Methanomassiliicoccus and Methanobacterium became the dominant archaea genera after MPW addition. 16S functional prediction analysis results indicated that genes expression of key enzymes involved in syntrophic acetate oxidation (SAO), hydrogenotrophic and methylotrophic methanogenesis were up-regulated, and acetoclastic methanogenesis was inhibited after MPW addition. Based on these analyses, it could be inferred that SAO combined with hydrogenotrophic and methylotrophic methanogenesis was the dominant pathway for organics degradation and methane production during AcD. These findings provided systematic insights into the microbial community changes and metabolic pathways during AcD of WAS and MPW.

Realization of 2D multiferroic with strong magnetoelectric coupling by intercalation: a first-principles high-throughput prediction

The discovery of novel two-dimensional (2D) multiferroic materials is attractive due to their potential for the realization of information storage and logic devices. Although many approaches have been explored to simultaneously introduce ferromagnetic (FM) and ferroelectric (FE) orders into a 2D material, the resulting systems are often plagued by weak magnetoelectric (ME) coupling or limited room-temperature stability. Here, we present a superlattice strategy to construct non-centrosymmetric AM2X4 multiferroic monolayers, i.e., intercalating transition metal ions (A) into the tetragonal-like vacancies of transition metal dichalcogenide bilayers (MX2). Starting from 960 intercalated AM2X4 compounds, our high-throughput calculations have identified 21 multiferroics with robust magnetic order, large FE polarization, low transition barrier, high FE/FM transition temperature, and strong ME coupling. According to the origin of magnetism, we have classified them into twelve type-a, seven type-b, and two type-c multiferroics, which exhibit different ME coupling behavior. During the switching of polarization, the reversal of skyrmions chirality, the transition of the magnetic ground state from FM to antiferromagnetic, and the changes in spin-polarized electron distribution were observed in type-a, type-b, and type-c 2D multiferroic materials, respectively. These results substantially expand the family of 2D ferroic materials and pave an avenue for designing and implementing nonvolatile logic and memory devices.

High-throughput data-driven machine learning prediction of thermal expansion coefficients of high-entropy solid solution carbides

Recent advances in machine learning and the expanding availability of materials data have enabled significant developments in materials science. In this study, novel configurations of high-entropy ceramic (HEC) materials were explored by predicting their coefficient of thermal expansion (CTE) using machine learning (ML) and high-throughput screening. A machine learning model was built using 3360 datasets containing the thermodynamic, elastic, and thermophysical properties of HEC with carbide configurations of (Ti0.2Ta0.2X0.2Y0.2Z0.2)C. The high correlation of the bulk and Young's moduli, and cohesive energy features with the CTE facilitated its prediction. The random forest (RF) and neural network (NET)-based models successfully reproduced the CTE reported in existing experimental and theoretical studies. Overall, first-principles calculation was implemented to configure a database for HEC and a new ML application method is proposed.

HemoDL: Hemolytic peptides prediction by double ensemble engines from Rich sequence-derived and transformer-enhanced information

Hemolytic peptides can trigger hemolysis by rupturing red blood cells’ membranes and triggering cell disruption. Due to the labor-intensive and time-consuming in-lab identification process, accurate, high-throughput hemolytic peptide prediction is crucial for the growth of peptide sequence data in proteomics and peptidomics. In this study, we offer the HemoDL ensemble learning model, which learns the distinct distribution of sequence characteristics for predicting the hemolytic activity of peptides using a double LightGBM framework. To determine the most informative encoding features, we compare 17 widely used features across four benchmark datasets. Our investigation reveals that CTD, BPF, Charge, AAC, GDPC, ATC, QSO, and transformer-based features exhibit more positive contributions to detecting the hemolytic activity of peptides. Comparison with eight state-of-the-art methods demonstrates that HemoDL outperforms other models, attaining higher Matthews Correlation Coefficient values on four test datasets, ranging from 6.30% to 16.04%, 6.63%–11.26%, 4.76%–9.92%, and 7.41%–15.03%, respectively. Additionally, we provide the HemoDL with a user-friendly graphical interface available at https://github.com/abcair/HemoDL. In summary, the HemoDL model, leveraging CTD, BPF, Charge, AAC, GDPC, ATC, QSO and transformer-based encoding features within a double LightGBM learning framework, achieves high accuracy in predicting the hemolytic activity of peptides.

High-throughput prediction of protein conformational distributions with subsampled AlphaFold2

This paper presents an innovative approach for predicting the relative populations of protein conformations using AlphaFold 2, an AI-powered method that has revolutionized biology by enabling the accurate prediction of protein structures. While AlphaFold 2 has shown exceptional accuracy and speed, it is designed to predict proteins’ ground state conformations and is limited in its ability to predict conformational landscapes. Here, we demonstrate how AlphaFold 2 can directly predict the relative populations of different protein conformations by subsampling multiple sequence alignments. We tested our method against nuclear magnetic resonance experiments on two proteins with drastically different amounts of available sequence data, Abl1 kinase and the granulocyte-macrophage colony-stimulating factor, and predicted changes in their relative state populations with more than 80% accuracy. Our subsampling approach worked best when used to qualitatively predict the effects of mutations or evolution on the conformational landscape and well-populated states of proteins. It thus offers a fast and cost-effective way to predict the relative populations of protein conformations at even single-point mutation resolution, making it a useful tool for pharmacology, analysis of experimental results, and predicting evolution.

Challenges in High-Throughput Inorganic Materials Prediction and Autonomous Synthesis

Challenges in High-Throughput Inorganic Materials Prediction and Autonomous Synthesis

Effects of microtopography on soil microbial communities in alpine meadows on the Qinghai-Tibetan Plateau

Heterogeneous habitats generated by diverse microtopography are critical in shaping soil microbial communities and their ecological functions. In this study, a combination of high-throughput sequencing technology and functional prediction was used to assess the effects of three microtopographic types (the shady slope, ravine, and sunny slope) on community structure, diversity, functional guilds, and the co-occurrence networks of soil microbes in alpine meadows. The results showed that the microbial α-diversity varied significantly across the three microtopographies, and fungal community structure was more responsive to microtopography than that of bacteria. In contrast, the co-occurrence network pattern of bacteria was more sensitive to environmental changes induced by microtopography. Mantel tests revealed that bacterial community structure was mainly impacted by plant diversity, belowground biomass, soil total phosphorus content and soil pH, and variations in fungal community structure were correlated with plant diversity, soil pH, and soil NH4+–N content. Meanwhile, regression analyses indicated that nitrogen fixers, ectomycorrhizal fungi, and endophytes relative abundances were positively linked to plant diversity, and soil pH negatively affected nitrogen fixers, ectomycorrhizal fungi, and methane oxidizers. Furthermore, different functional guilds with similar functions (e.g., ectomycorrhizal fungi vs. arbuscular mycorrhizal fungi) responded inconsistently to the microtopography, suggesting ecological niche differentiation of microorganisms in microtopographies. These findings highlight the importance of microtopography for driving bacterial and fungal diversity, community structure, functional profiles and co-occurrence networks in alpine ecosystems.

QSPR modeling for the prediction of partitioning of VOCs and SVOCs to indoor fabrics: Integrating environmental factors

Porous fabrics have a significant impact on indoor air quality by adsorbing and emitting chemical substances, such as volatile organic compounds (VOCs) and semi-volatile organic compounds (SVOCs). Understanding the partition behavior between organic compound molecules and indoor fabrics is crucial for assessing their environmental fate and associated human exposure. The physicochemical properties of fabrics and compounds are fundamental in determining the free energy of partitioning. Moreover, environmental factors like temperature and humidity critically affect the partition process by modifying the thermal and moisture conditions of the fabric. However, existing methods for determining the fabric-air partition coefficient are limited to specific fabric-chemical combinations and lack a comprehensive consideration of indoor environmental factors. In this study, large amounts of experimental data on fabric-air partition coefficients (Kfa) of (S)VOCs were collected for silk, polyester, and cotton fabrics. Key molecular descriptors were identified, integrating the influences of physicochemical properties, temperature, and humidity. Subsequently, two typical quantitative structure-property relationship (QSPR) models were developed to correlate the Kfa values with the molecular descriptors. The fitting performance, robustness, and predictive ability of the two QSPR models were evaluated through statistical analysis and internal/external validation. This research provides insights for the high-throughput prediction of the environmental behaviors of indoor organic compounds.

Identification of Active Components for Sports Supplements: Machine Learning-Driven Classification and Cell-Based Validation.

The identification of active components is critical for the development of sports supplements. However, high-throughput screening of active components remains a challenge. This study sought to construct prediction models to screen active components from herbal medicines via machine learning and validate the screening by using cell-based assays. The six constructed models had an accuracy of >0.88. Twelve randomly selected active components from the screening were tested for their active potency on C2C12 cells, and 11 components induced a significant increase in myotube diameters and protein synthesis. The effect and mechanism of luteolin among the 11 active components as potential sports supplements were then investigated by using immunofluorescence staining and high-content imaging analysis. It showed that luteolin increased the skeletal muscle performance via the activation of PGC-1α and MAPK signaling pathways. Thus, high-throughput prediction models can be effectively used to screen active components as sports supplements.

Towards the high-throughput prediction of finite-temperature properties using the quasi-harmonic approximation

Lattice dynamics calculations within the quasi-harmonic approximation (QHA) provide an infrastructure for modelling the finite-temperature properties of periodic solids at a modest computational cost. With the recent widespread interest in materials discovery by data mining, a database of computed finite-temperature properties would be highly desirable. In this work we provide a first step toward this goal with a comparative study of the accuracy of five exchange-correlation functionals, spanning the local density approximation (LDA), generalised-gradient approximation (GGA) and meta-GGA levels of theory, for predicting the properties of ten Group 1, 2 and 12 binary metal oxides. We find that the predictions are bounded by the LDA, which tends to underestimate lattice parameters and cell volumes relative to experiments, but yields the most accurate results for bulk moduli, expansion coefficients and Grüneisen parameters, and the PBE GGA, which shows the opposite behaviour. The PBEsol GGA gives the best overall predictions of the lattice parameters and volumes whilst also giving relatively reliable results for other properties. Our results demonstrate that, given a suitable choice of functional, a variety of finite-temperature properties can be predicted with useful accuracy, and hence that high-throughout QHA calculations are technically feasible.

Mixed Enthalpy-Entropy Descriptor for the Rational Design of Synthesizable High-Entropy Materials Over Vast Chemical Spaces.

The practically unlimited high-dimensional composition space of high-entropy materials (HEMs) has emerged as an exciting platform for functional material design and discovery. However, the identification of stable and synthesizable HEMs and robust design rules remains a daunting challenge. Here, we propose a mixed enthalpy-entropy descriptor (MEED) that enables highly efficient, robust, high-throughput prediction of synthesizable HEMs across vast chemical spaces from first-principles. The MEED is based on two parameters: the relative formation enthalpy with respect to the most stable competing compound and the spread of the point-defect formation energy spectrum. The former measures the relative synthesizability of an HEM to its most stable competing phase, going beyond the conventional thermodynamic understanding. The latter gauges the relative entropy forming ability of an HEM, entailing no sampling over numerous alloy configurations. By applying the MEED to two structurally distinct representative material systems (i.e., 3D rocksalt carbides and 2D layered sulfides), we not only successfully identify all experimentally reported HEMs within these systems but also reveal a cutoff criterion for assessing their relative synthesizability within each system. By the MEED, tens of new high-entropy carbides and 2D high-entropy sulfides are also predicted, which have the potential for a wide variety of applications such as coating in aerospace devices, energy conversion and storage, and flexible electronics.