Search

Abstract Pretrained Graph Neural Networks have been widely adopted for various molecular property prediction tasks. Despite their ability to encode structural and relational features of molecules, traditional fine-tuning of such pretrained GNNs on the target task can lead to poor generalization. To address this, we explore the adaptation of pretrained GNNs to the target task by jointly training them with multiple auxiliary tasks. This could enable the GNNs to learn both general and task-specific features, which may benefit the target task. However, a major challenge is to determine the relatedness of auxiliary tasks with the target task. To address this, we investigate multiple strategies to measure the relevance of auxiliary tasks and integrate such tasks by adaptively combining task gradients or by learning task weights via bi-level optimization. Additionally, we propose a novel gradient surgery-based approach, Rotation of Conflicting Gradients ( $$\mathop {\texttt{RCGrad}}\limits$$ RCGrad ), that learns to align conflicting auxiliary task gradients through rotation. Our experiments with state-of-the-art pretrained GNNs demonstrate the efficacy of our proposed methods, with improvements of up to 7.7% over fine-tuning. This suggests that incorporating auxiliary tasks along with target task fine-tuning can be an effective way to improve the generalizability of pretrained GNNs for molecular property prediction. Scientific contribution We introduce a novel framework for adapting pretrained GNNs to molecular tasks using auxiliary learning to address the critical issue of negative transfer. Leveraging novel gradient surgery techniques such as $$\mathop {\texttt{RCGrad}}\limits$$ RCGrad , the proposed adaptation framework represents a significant departure from the dominant pretraining fine-tuning approach for molecular GNNs. Our contributions are significant for drug discovery research, especially for tasks with limited data, filling a notable gap in the efficient adaptation of pretrained models for molecular GNNs.

Objective Riveting quality issues cannot be visually judged. By employing the far-field eddy current testing method, internal quality defects in riveted structural components can be detected, providing theoretical and technical support for product quality inspection. Method Riveting quality inspection and classification of clearance types are conducted without damaging the workpiece. Far-field eddy current testing experiments are carried out for the intermediate plate clearance and rivet hole clearance of rail vehicle riveted structural components respectively. Testing research on metro vehicle anti-roll torsion bar seats riveted to the body frame side beams is also conducted. Result & Conclusion From the results of the three sets of experiments, it is evident that the range of phase angles can distinguish between rivet and hole clearances and plate clearances. The phase amplitude range for hole clearances is 60° to 90°, while for plate clearances, it is 0 to 10°.

Abstract Due to cancer’s complex nature and variable response to therapy, precision oncology informed by omics sequence analysis has become the current standard of care. However, the amount of data produced for each patient makes it difficult to quickly identify the best treatment regimen. Moreover, limited data availability has hindered computational methods’ abilities to learn patterns associated with effective drug-cell line pairs. In this work, we propose the use of contrastive learning to improve learned drug and cell line representations by preserving relationship structures associated with drug mechanisms of action and cell line cancer types. In addition to achieving enhanced performance relative to a state-of-the-art method, we find that classifiers using our learned representations exhibit a more balanced reliance on drug- and cell line-derived features when making predictions. This facilitates more personalized drug prioritizations that are informed by signals related to drug resistance.

The developmental toxicity and human health risks of triazole fungicides (TFs) have attracted worldwide attention due to the ability to enter the human body in a variety of ways. Nevertheless, the specific mechanism by which TFs exert remains incompletely understood. Given that retinoic acid (RA) signaling pathway are closely related to development, this study aimed to screen and identify developmentally disabled chemicals in commonly used TFs and to reveal the potential effects of TFs on developmental retardation through the RA signaling pathway in mouse embryonic stem cells (mESCs). Specifically, six typical TFs (myclobutanil, tebuconazole, hexaconazole, propiconazole, difenoconazole, and flusilazole) were exposed through the construction of an embryoid bodies (EBs)-based in vitro global differentiation models. Our results clarified that various TFs disturbed lineage commitment during early embryonic development. Crucially, the activation of RA signaling pathway, which alters the expression of key genes and interferes the transport and metabolism of retinol, may be responsible for this effect. Furthermore, molecular docking, molecular dynamics simulations, and experiments using a retinoic acid receptor α inhibitor provide evidence supporting the potential modulatory role of the retinoic acid signaling pathway in developmental injury. The current study offers new insights into the TFs involved in the RA signaling pathway that interfere with the differentiation process of mESCs, which is crucial for understanding the impact of TFs on pregnancy and early development.

ObjectiveExisting literature has not clearly elucidated whether SARS-CoV-2 infection increases the incidence of Parkinson’s disease or if Parkinson’s disease patients are more susceptible to the effects of SARS-CoV-2 infection. To clarify the issue, this study employs a genetic epidemiological approach to investigate the association.MethodsThis study utilizes a two-sample Mendelian randomization analysis. The primary analysis employs the inverse variance-weighted (IVW) method, supplemented by secondary analyses including MR-Egger regression, weighted median, IVW radial method, and weighted mode, to evaluate the bidirectional causal relationship between Parkinson’s disease and SARS-CoV-2 infection.ResultsIVW results showed no genetic causality between SARS-CoV-2 susceptibility, hospitalization rate and severity and Parkinson’s disease. (IVW method: p = 0.408 OR = 1.10 95% CI: 0.87 ~ 1.39; p = 0.744 OR = 1.11 95% CI: 0.94 ~ 1.09; p = 0.436 OR = 1.05 95% CI: 0.93 ~ 1.17). Parkinson’s disease was not genetically associated with susceptibility to new crown infections, hospitalization rates, and severity (IVW method: p = 0.173 OR = 1.01 95% CI: 0.99 ~ 1.03; p = 0.109 OR = 1.05 95% CI: 0.99 ~ 1.12; p = 0.209 OR = 1.03 95% CI: 0.99 ~ 1.07). MR-Egger regression, weighted median, IVW radial method, and weighted mode results are consistent with the results of the IVW method.ConclusionThis study does not support a genetic link between Parkinson’s disease and SARS-CoV-2 infection, and the association observed in previous cohort studies and observational studies may be due to other confounding factors.

Odor compounds produced in salted egg yolk during thermal processing were identify by sensory evaluation and gas chromatography-mass spectrometry (GC-MS) combined with multivariate statistical analysis such as hierarchical cluster analysis (HCA), orthogonal partial least squares-discriminant analysis (OPLS-DA) and principal component analysis (PCA). By using quantitative descriptive sensory analysis, it was found that there was a significant difference in the flavor profile between egg yolk samples with and without unpleasant odor, and it was determined that some egg yolks had obvious off-flavor during hot processing. Cluster analysis showed that the six groups of salted egg yolk samples were clearly divided into two categories, which was consistent with the sensory evaluation results. Furthermore, OPLS-DA identified 33 major differential components with variable importance in the projection (VIP) scores greater than one between samples with and without unpleasant odor, and aldehydes and alcohols were the most significant differential components. Finally, the volatile compounds showing a positive correlation with sour odor were screened by PCA, and n-hexanal, n-valeraldehyde, 1-octen-3-ol, trans,trans-2,4-decadienal, styrene, dimethyl disulfide and 2-nonone were identified due to their odor activity values (OAV) greater than one as the major odor compounds of salted egg yolk during thermal processing. The results provide a reference for analysis of the flavor characteristics and possible off-flavor components of salted egg yolk after thermal processing and provide a scientific basis for accessing the quality of salted egg yolk.

To characterize the quality of widely consumed pork in China, the chemical compositions and other indexes of five breeds of pork were compared. The results indicated that the moisture content, sensory flavor, and overall acceptability of Pipa pork (PPP) were superior to other breeds. The fat content and essential amino acids (EAA) of Yihao native pork (YNP) were significantly (p < 0.05) higher than in other breeds. The protein content, the total amount of amino acids, and perceptible flavor of Tibetan black pork (TBP) were higher than in other breeds. The protein nutrition profiles of manor black pork (MBP) and TBP were better than in others. The equivalent umami concentration (EUC) value of white pork (WP) was significantly (p < 0.05) higher than in others; however, the health risk of its fatty acid content was higher. There were unique protein and flavor chemicals in YNP, TBP, and PPP, which may be useful for distinguishing their authenticity.