Motivation: Identifying drug–target interactions (DTI) is crucial in drug discovery. Fragments are less complex and can accurately characterize local features, which is important in DTI prediction. Recently, deep learning (DL)-based methods predict DTI more efficiently. However, two challenges remain in existing DL-based methods: (i) some methods directly encode drugs and proteins into integers, ignoring the substructure representation; (ii) some methods learn the features of the drugs and proteins separately instead of considering their interactions. Results: In this article, we propose a fragment-oriented method based on a multihead cross attention mechanism for predicting DTI, named FMCA-DTI. FMCA-DTI obtains multiple types of fragments of drugs and proteins by branch chain mining and category fragment mining. Importantly, FMCA-DTI utilizes the shared-weight-based multihead cross attention mechanism to learn the complex interaction features between different fragments. Experiments on three benchmark datasets show that FMCA-DTI achieves significantly improved performance by comparing it with four state-of-the-art baselines. Availability and implementation: The code for this workflow is available at: https://github.com/jacky102022/FMCA-DTI.
14 August 2023 in Environmental Science and Pollution Research
Most of the methods for predicting air pollutant concentrations are targeting at single pollutants, which is time-consuming and laborious. To solve this problem, this study proposes a Convolutional Transformer (Convtrans) model that incorporates knowledge to make a collaborative estimation of PM2.5 and O3 by combining ground, satellite, and reanalysis data. Knowledge is introduced into the model by the shared and specific inputs, the PM2.5-O3 interaction module, and the weighted loss function designed with the correlation between PM2.5 and O3 concentrations. To verify the accuracy of the Convtrans model, its prediction result was compared with that of CNN-LSTM, Transformer, RF, and XGB models. Estimating the pollutant concentration in typical Chinese cities, the cross-validation results show that Convtrans has the minimum error (PM2.5:RMSE = 6.136 µg/m³, O3:RMSE = 8.250 µg/m³) and the highest prediction accuracy (PM2.5:R2 = 0.923, O3:R2 = 0.898). Finally, a map of pollutant concentrations was drawn according to the pollutant concentration values predicted by the model, showing the spatial variations of pollutant. This study indicates that it is feasible to integrate knowledge into a data-driven model for a joint estimation of atmospheric pollutant concentrations. In addition, the joint estimation framework for pollutants proposed in this study can be applied to multivariate retrieval or estimation in multiple fields.
Paired spatial multi-omics provides a supervised basis for learning RNA–protein correspondence in situ, but predicting protein abundance from spatial transcriptomic data alone remains challenging across tissue contexts and protein panels. Here, we present DPAS-Graph, an adaptive relation-learning framework for spatial RNA-to-protein prediction. Rather than directly merging spatial proximity and transcriptomic similarity as fixed graph priors, DPAS-Graph represents them as two relation channels on a shared edge support and updates their contributions during representation learning for protein prediction. Its Niche-Coupled Field Encoder combines layer-wise edge-relation modeling, intra-branch relation refinement, and cross-branch residual correction to learn spot representations for protein abundance prediction. In a leave-one-dataset-out benchmark across seven paired spatial multi-omics datasets, DPAS-Graph achieved lower aggregate prediction errors and improved spot-level agreement of protein expression profiles, with gains mainly reflected in error-based metrics and PCC-Spot. Spatial autocorrelation and protein-derived domain agreement analyses were further used to characterize the spatial behavior of the predicted protein maps. When applied to external RNA-only spatial sections, DPAS-Graph generated qualitatively interpretable marker-level virtual protein maps, illustrating its use as a complementary tool for protein-level interpretation of transcriptomics-only spatial data.