Majority Of Predictions Research Articles

CKG-IMC: An inductive matrix completion method enhanced by CKG and GNN for Alzheimer’s disease compound-protein interactions prediction

Alzheimer’s disease (AD) is one of the most prevalent chronic neurodegenerative disorders globally, with a rapidly growing population of AD patients and currently no effective therapeutic interventions available. Consequently, the development of therapeutic anti-AD drugs and the identification of AD targets represent one of the most urgent tasks. In this study, in addition to considering known drugs and targets, we explore compound-protein interactions (CPIs) between compounds and proteins relevant to AD. We propose a deep learning model called CKG-IMC to predict Alzheimer’s disease compound-protein interaction relationships. CKG-IMC comprises three modules: a collaborative knowledge graph (CKG), a principal neighborhood aggregation graph neural network (PNA), and an inductive matrix completion (IMC). The collaborative knowledge graph is used to learn semantic associations between entities, PNA is employed to extract structural features of the relationship network, and IMC is utilized for CPIs prediction. Compared with a total of 16 baseline models based on similarities, knowledge graphs, and graph neural networks, our model achieves state-of-the-art performance in experiments of 10-fold cross-validation and independent test. Furthermore, we use CKG-IMC to predict compounds interacting with two confirmed AD targets, 42-amino-acid β-amyloid (Aβ42) protein and microtubule-associated protein tau (tau protein), as well as proteins interacting with five FDA-approved anti-AD drugs. The results indicate that the majority of predictions are supported by literature, and molecular docking experiments demonstrate a strong affinity between the predicted compounds and targets.

Predicting and screening high-performance polyimide membranes using negative correlation based deep ensemble methods.

Polyimide polymer membranes have become critical materials in gas separation and storage applications due to their high selectivity and excellent permeability. However, with over 107 known types of polyimides, relying solely on experimental research means potential high-performance candidates are likely to be overlooked. This study employs a deep learning method optimized by negative correlation ensemble techniques to predict the gas permeability and selectivity of polyimide structures, enabling rapid and efficient material screening. We propose a deep neural network model based on negative correlation deep ensemble methods (DNN-NCL), using Morgan molecular fingerprints as input. The DNN-NCL model achieves an R2 value of approximately 0.95 on the test set, which is a 4% improvement over recent model performance, and effectively mitigates overfitting with a maximum discrepancy of less than 0.03 between the training and test sets. High-throughput screening of over 8 million hypothetical polymers identified hundreds of promising candidates for gas separation membranes, with 14 structures exceeding the Robeson upper bound for CO2/N2 separation. Visualization of high-throughput predictions shows that although the Robeson upper bound was never explicitly used as a model constraint, the majority of predictions are compressed below this limit, demonstrating the deep learning model's ability to reflect real-world physical conditions. Reverse analysis of model predictions using SHAP analysis achieved interpretability of the deep learning model's predictions and identified three key functional groups deemed important by the deep neural network for gas permeability: carbonyl, thiophene, and ester groups. This established a bridge between the structure and properties of polyimide materials. Additionally, we confirmed that two polyimide structures predicted by the model to have excellent CO2/N2 selectivity, namely 6-methylpyrimidin-5-amine and 1,4,5,6-tetrahydropyrimidin-2-amine, have been experimentally validated in previous studies. This research demonstrates the feasibility of using deep learning methods to explore the vast chemical space of polyimides, providing a powerful tool for discovering high-performance gas separation membranes.

Identification of single nucleotide variants in SLC26A9 gene in patients with cystic fibrosis (p.Phe508del homozygous) and its association to Orkambi® (Lumacaftor and Ivacaftor) response in vitro

BackgroundSince patients with cystic fibrosis with different Cystic Fibrosis Transmembrane Regulator (CFTR) genotypes present a wide response variability for modulator drugs such as Orkambi®, it is important to screen variants in candidate genes with an impact on precision and personalized medicine, such as Solute Carrier Family 26, member 9 (SLC26A9) gene. MethodsSanger sequencing for the exons and intron–exon boundary junctions of the SLC26A9 gene was employed in nine individuals with p.Phe508del homozygous genotype for the CFTR gene who were not under CFTR modulators therapy. The sequencing variants were evaluated by in silico prediction tools. The CFTR function was measured by cAMP–stimulated current (ΔIsc-eq-FSK) in polarized CFTR of human nasal epithelial cells cultured in micro–Ussing chambers with Orkambi®. ResultsWe found 24 intronic variants, three in the coding region (missense variants – rs74146719 and rs16856462 and synonymous – rs33943971), and three in the three prime untranslated region (3′ UTR) region in the SLC26A9 gene. Twenty variants were considered benign according to American College of Medical Genetics and Genomics guidelines, and ten were classified as uncertain significance. Although some variants had deleterious predictions or possible alterations in splicing, the majority of predictions were benign or neutral. When we analyzed the ΔIsc-eq-FSK response to Orkambi®, there were no significant differences within the genotypes and alleles for all 30 variants in the SLC26A9 gene. ConclusionsAmong the nine individuals with p.Phe508del homozygous genotype for the CFTR gene, no pathogenic SLC26A9 variants were found, and we did not detect associations from the 30 SLC26A9 variants and the response to the Orkambi® in vitro.

Performance of four models for eye color prediction in an Italian population sample

Forensic DNA phenotyping (FDP) has recently provided important advancements in forensic investigations, by predicting the physical appearance of a subject from a biological sample, using SNP markers. The majority of operable prediction models have been developed for iris color; however, replication studies to understand their applicability on a worldwide scale are still limited for many of them.In this work, 4 models for eye color prediction (IrisPlex, Ruiz, Allwood and Hart models) were systematically evaluated in a sample of 296 subjects of Italian origin. Genotypes were determined by a custom NGS-based panel targeting all the predictive SNPs included in the 4 tested models.Overall, 60–69% of the Italian sample could be correctly predicted with the IrisPlex, Ruiz and Allwood models, applying the recommended threshold. The IrisPlex model showed the lowest frequency of errors (17%), but also the highest number of inconclusive results (18%). In the absence of the threshold, the highest proportion of correct predictions was again obtained with the IrisPlex model (76%), followed by the Allwood (73%) and the Ruiz (65%) models. Lastly, the Hart predictive algorithm had the lowest error rate (2%), but the majority of predictions (87%) were restricted to the less informative categories of “not-blue” and “not-brown”, and correct color predictions were obtained only for 11% of the sample.As observed in previous studies, the majority of incorrect and undefined predictions were ascribable to the intermediate category, which represented 25% of the Italian sample. An adjustment of the IrisPlex (multinomial logistic regression) and Ruiz models (Snipper Bayesian classifier) with Italian allele frequencies gave only minor improvements in predicting intermediate eye color and no remarkable overall changes in performance. This suggests an incomplete knowledge underlying the intermediate colors. Considering the impact of this phenotype in the Italian sample as well as in other admixed populations, future improvements of eye color prediction methods should include a better genetic and phenotypic characterization of this category.

Forecasting the unemployment rate over districts with the use of distinct methods

Abstract Interdependencies among neighboring regions appear to be important in forming the shape of local labor markets. Nevertheless, only a few studies exist which have applied spatial models to forecast over small spatial units such as cities, districts or counties. The majority of predictions are developed with quarterly or yearly time series for a country or at regional level. The paper presents the above phenomena and deals with the problem of simultaneous forecasting of the unemployment rate over 35 poviats (districts and cities) in one of the Polish provinces. Two extremely different models with spatial dependencies were developed and estimated in this paper: the Spatial Vector Autoregressions (SpVAR) and the Spatial Artificial Neural Network (SpANN). The 13-month out-of-sample forecast is based on high frequency, raw, monthly panel data extracted from 31 local labor offices. The procedure worked out here allows comparing the forecasting performance of spatial models with their non-spatial and seasonal equivalents. The inclusion of a spatial component into the models significantly improves the accuracy of forecasts; however, the overall performance of SpVAR is 30% better than SpANN.

Past and future impact of climate change on foraging habitat suitability in a high-alpine bird species: Management options to buffer against global warming effects

The majority of predictions about the impacts of climate change on wildlife have relied either on the study of species' physiological tolerance or on broad-scale distribution models. In comparison, little attention has been paid to species' mechanistic responses to fine-grained, climate-induced modifications of habitat suitability. However, such studies would be pivotal to the understanding of species' ecological requirements (and hence their adaptive potential to environmental change) and the design of management strategies. We investigated foraging microhabitat selection in a potentially climate-change sensitive species, the white-winged snowfinch Montifringilla nivalis, during the breeding season in the Alps. Our microhabitat selection model considered topography, ground-cover variables and sward height within a 5-m radius at foraging and control locations. Habitat selection was positively affected by grassland cover, negatively by sward height and quadratically by snow cover (optimum around 40%); birds avoided anthropized (urban areas, roads) sites. We estimated past (1976) and future (2066) climate-driven changes in foraging microhabitat suitability, assuming a progressively earlier date of snowmelt due to increasing temperatures over this entire time span. We then modelled the potential impact of snowmelt (and related sward height) on habitat suitability under two scenarios: maintaining the current situation (i.e. irregular seasonal grazing) and implementing targeted management in an attempt to mitigate impacts of earlier snowmelt. Predicted foraging habitat suitability (estimated as the fraction of suitable plots) significantly declined over time (−23% between 1976 and 2016, further 32% loss by 2066). However, model outputs demonstrated that maintaining sward height below 6 cm on breeding grounds (e.g. by regular grazing) would significantly decrease the predicted loss of suitable foraging habitat. Detailed information about patterns of resource exploitation allows the identification of mechanistic, functional responses of species to environmental change, and enables an evaluation of habitat management options that can buffer against the detrimental effects of global warming.

Optimization of the octane response of gasoline/ethanol blends

The octane responses of gasoline/ethanol mixtures are not well understood because of the unidentified intermolecular interactions in such blends. In general, when ethanol is blended with gasoline, the Research Octane Number (RON) and the Motor Octane Number (MON) non-linearly increase or decrease, and the non-linearity is determined by the composition of the base gasoline and the amount of added ethanol. The complexity of commercial gasolines, comprising of hundreds of different components, makes it challenging to understand ethanol-gasoline synergistic/antagonistic blending effects. Understanding ethanol blending effects with simpler gasoline surrogates is critical to acquire knowledge about ethanol blending with complex multi-component gasoline fuels. In this study, the octane numbers (ON) of ethanol blends with five relevant gasoline surrogate molecules were measured. The molecules investigated in this study include: n-pentane, iso-pentane, 1,2,4-trimethylbenzene, cyclopentane and 1-hexene. These new measurements along with the available data of n-heptane, iso-octane, toluene, various primary reference fuels (PRF) and toluene primary reference fuels (TPRF) with ethanol are used to develop a blending rule for the octane response (RON and MON) of multi-component blends with ethanol. In addition, new ON data are collected for six Fuels for Advanced Combustion Engine (FACE) with ethanol. The relatively simple volume based model successfully predicts the octane numbers (ON) of the various ethanol/PRF and ethanol/TPRF blends with the majority of predictions being within the ASTM D2699 (RON) and D2700 (MON) reproducibility limits. The model is also successfully validated against the ON of the FACE gasolines blended with ethanol with the majority of predictions being within the reproducibility limits. Finally, insights into the possible causes of the synergistic and antagonistic effects of different molecules with ethanol are provided.

Measurement of top quark pair differential cross sections in the dilepton channel in pp collisions at s=7 and 8 TeV with ATLAS

Measurements of normalized differential cross-sections of top quark pair ($t\bar t$) production are presented as a function of the mass, the transverse momentum and the rapidity of the $t\bar t$ system in proton-proton collisions at center-of-mass energies of $\sqrt{s}$ = 7 TeV and 8 TeV. The dataset corresponds to an integrated luminosity of 4.6 fb$^{-1}$ at 7 TeV and 20.2 fb$^{-1}$ at 8 TeV, recorded with the ATLAS detector at the Large Hadron Collider. Events with top quark pair signatures are selected in the dilepton final state, requiring exactly two charged leptons and at least two jets with at least one of the jets identified as likely to contain a $b$-hadron. The measured distributions are corrected for detector effects and selection efficiency to cross-sections at the parton level. The differential cross-sections are compared with different Monte Carlo generators and theoretical calculations of $t\bar t$ production. The results are consistent with the majority of predictions in a wide kinematic range.

A blending rule for octane numbers of PRFs and TPRFs with ethanol

Ethanol is widely used as an octane booster in commercial gasoline fuels. Its oxygenated nature aids in reducing harmful emissions such as nitric oxides (NOx), soot and unburned hydrocarbons (HC). However, the non-linear octane response of ethanol blending with gasoline fuels is not completely understood because of the unknown intermolecular interactions in such blends. In general, when ethanol is blended with gasoline, the Research Octane Number (RON) and the Motor Octane Number (MON) non-linearly increase (synergistic) or decrease (antagonistic), and the non-linearity depends on the composition of the base gasoline. The complexity of commercial gasoline, comprising of hundreds of different components, makes it challenging to understand ethanol–gasoline synergistic/antagonistic blending effects. Understanding ethanol blending effects with simpler gasoline surrogates blends may enable a better understanding of ethanol blending with complex multi-component gasoline fuels. This study presents a blending rule to predict the octane numbers (ON) of ethanol/primary reference fuel (PRF; mixtures of iso-octane and n-heptane) and ethanol/toluene primary reference fuel (TPRF; mixtures of toluene, iso-octane and n-heptane) mixtures using the data available in literature and new data. The ON of ethanol blends with PRF-40, -50, and -60 were measured and compared with those from literature. Additional experimental data were collected to validate the developed model for ethanol blends of three different TPRFs having the same RON but different MON (i.e., different toluene contents). The three tested TPRF mixtures have octane ratings of RON 60.0/MON 58.0 (toluene 10.2vol%), RON 60.0/MON 56.3 (toluene 19.8vol%), and RON 60.0/MON 53.2 (toluene 40.2vol%). The octane prediction model consists of linear and non-linear by mole regions. The transition point between the linear and non-linear regions is a function of the RON and MON of the base PRF and TPRF mixture. The non-linear by mole region is defined as a second-order polynomial and the combined model successfully predicts the octane number (ON) of the various ethanol/PRF and ethanol/TPRF blends with a maximum ON error of 2.7, with the majority of predictions being within the reproducibility limits. Finally, the model presented here is shown to be an improvement over that existing already in the literature.

L1 vowels of multilinguals: the applicability of SLM in multilingualism

Although L1 has been treated as a rigid system which is more likely to act as a sender than a receiver of CLI in bilinguals and multilinguals, recent studies have provided some evidence of the influence of both L2 and L3 on L1. The study is aimed at shedding further light on how Lns can influence the native language and how these changes can be explained by means of the Speech Learning Model. The first and second formant of L1 Polish vowels of three groups of multilinguals were compared. Evidence of a systemic influence of L2 on L1 was observed in the raising and backing of L1 Polish vowels due to L2 English and lowering and backing or fronting of L1 Polish vowels due to L2 German. No systemic influence of L3 on L1 was observed. The predictions derived from equivalence classification of SLM were tested for the Polish vowel /ɛ/ and the closest vowels from Lns. The majority of predictions regarding the convergence or divergence of the particular diaphone were supported by the data.

The Anxiety Attitude and Belief Scale-2: Development, Measurement Model, and Initial Validity.

There is a notable lack of measures of enduring beliefs, which are key etiological factors in Beck's cognitive model of anxiety. The Anxiety Attitude and Belief Scale-2 was developed to address this need. Items from the original AABS were reviewed and revised, and additional items were added to cover the range of constructs identified as reflecting anxiety related expectancies while avoiding the confounding of cognition and affect. Suitability of items was examined using cognitive interviewing (Willis, 2004). The resulting set of 48 items was administered to an index sample of individuals reporting anxiety symptoms and a cross-validation sample of undergraduate students in order to derive a measurement model describing its internal structure. The final, 33-item AABS-2 had a bifactor structure of one general and four specific factors, good fit to the data, common factor content across groups, acceptable precision in measurement, and evidence of construct validity. Measures of enduring beliefs related to anxiety disorders are needed to assess etiological factors within cognitive therapy; while there are numerous measures of automatic thoughts, there are few measures of beliefs. The present study sought to address this gap. The items that originally appeared on ten rationally derived scales drawn from clinical phenomenology of anxiety disorders were eventually grouped into four group factors and one general factor in the course of psychometric analyses. The group factors included ones expected to distinguish groups reporting panic, OCD, and social anxiety symptoms from other anxiety symptom groups, and this prediction was supported. The majority of predictions regarding patterns or correlations were also supported. Further validation research is needed to evaluate the validity of the AABS and its subscales in predicting course and outcome of psychotherapy.

Raptor displacement due to the construction of a wind farm: preliminary results after the first 2 years since the construction

Wind energy has experienced a very rapid growth in the last few years and the majority of predictions speculate a similar scenario also for the near future. Despite the fact that wind energy is carbon free, and thus does not contribute to global warming, wind farms are not free from impacting the environment in which they are built, particularly bird and bat populations – impacts that, at least in some cases, could be remarkable and thus need to be studied and monitored. In this paper, we present the results after the first year of monitoring, aiming to understand impacts of a wind farm on birds, particularly on raptors, both residents and migrators. The study area, located in central Tuscany, hosts a very rich raptor community with some species of conservation concern. Comparing data gathered before and after the wind farm construction, we note a dramatic decrease in the number of raptor observations, even though no species have definitely left the area. Significant differences arise also considering habitat use and bird behaviour, with raptors flying higher and away from turbines, stressing an avoidance effect.

An Evolutionary Psychological Investigation of Parental Distress and Reproductive Coercion During the “Coming Out” of Gay Sons

The lack of success of the “coming out” studies over the last three decades to explain and predict parental responses has motivated an evolutionary psychological reconceptualization. According to this reconceptualization, it was predicted that (a) biological mothers would experience more distress and apply more pressure on gay sons to change than would biological fathers and; (b) obligate investment for fathers on dependent sons would cause fathers to experience more distress and apply more pressure on gay sons to change than it would fathers without this obligate investment. In contrast, a cultural-norm hypothesis predicted that fathers would experience more distress and apply more pressure on gay sons to change than mothers. The majority of predictions were tested using 787 participants from two-biological parent families, who were drawn from a total sample of 891 participants from various family backgrounds. As predicted by the evolutionary hypothesis, biological mothers were reported to have been more distressed and coercive than biological fathers, in spite of a strong, societal expectation to the contrary. Furthermore, the results supported the obligate investment argument for paternal reactions. The model not only correctly explained and predicted parental behavior during coming out, but also was shown to unify within its theoretical framework discrepant results from the literature previously considered inconsistent.

COMPONENTS OF UNCERTAINTY IN SPECIES DISTRIBUTION ANALYSIS: A CASE STUDY OF THE GREAT GREY SHRIKE

Sophisticated statistical analyses are common in ecological research, particularly in species distribution modeling. The effects of sometimes arbitrary decisions during the modeling procedure on the final outcome are difficult to assess, and to date are largely unexplored. We conducted an analysis quantifying the contribution of uncertainty in each step during the model-building sequence to variation in model validity and climate change projection uncertainty. Our study system was the distribution of the Great Grey Shrike in the German federal state of Saxony. For each of four steps (data quality, collinearity method, model type, and variable selection), we ran three different options in a factorial experiment, leading to 81 different model approaches. Each was subjected to a fivefold cross-validation, measuring area under curve (AUC) to assess model quality. Next, we used three climate change scenarios times three precipitation realizations to project future distributions from each model, yielding 729 projections. Again, we analyzed which step introduced most variability (the four model-building steps plus the two scenario steps) into predicted species prevalences by the year 2050. Predicted prevalences ranged from a factor of 0.2 to a factor of 10 of present prevalence, with the majority of predictions between 1.1 and 4.2 (inter-quartile range). We found that model type and data quality dominated this analysis. In particular, artificial neural networks yielded low cross-validation robustness and gave very conservative climate change predictions. Generalized linear and additive models were very similar in quality and predictions, and superior to neural networks. Variations in scenarios and realizations had very little effect, due to the small spatial extent of the study region and its relatively small range of climatic conditions. We conclude that, for climate projections, model type and data quality were the most influential factors. Since comparison of model types has received good coverage in the ecological literature, effects of data quality should now come under more scrutiny.