Interaction-Based Inductive Bias in Graph Neural Networks: Enhancing Protein-Ligand Binding Affinity Predictions From 3D Structures.

Inductive bias in machine learning (ML) is the set of assumptions describing how a model makes predictions. Different ML-based methods for protein-ligand binding affinity (PLA) prediction have different inductive biases, leading to different levels of generalization capability and interpretability. Intuitively, the inductive bias of an ML-based model for PLA prediction should fit in with biological mechanisms relevant for binding to achieve good predictions with meaningful reasons. To this end, we propose an interaction-based inductive bias to restrict neural networks to functions relevant for binding with two assumptions: (1) A protein-ligand complex can be naturally expressed as a heterogeneous graph with covalent and non-covalent interactions; (2) The predicted PLA is the sum of pairwise atom-atom affinities determined by non-covalent interactions. The interaction-based inductive bias is embodied by an explainable heterogeneous interaction graph neural network (EHIGN) for explicitly modeling pairwise atom-atom interactions to predict PLA from 3D structures. Extensive experiments demonstrate that EHIGN achieves better generalization capability than other state-of-the-art ML-based baselines in PLA prediction and structure-based virtual screening. More importantly, comprehensive analyses of distance-affinity, pose-affinity, and substructure-affinity relations suggest that the interaction-based inductive bias can guide the model to learn atomic interactions that are consistent with physical reality. As a case study to demonstrate practical usefulness, our method is tested for predicting the efficacy of Nirmatrelvir against SARS-CoV-2 variants. EHIGN successfully recognizes the changes in the efficacy of Nirmatrelvir for different SARS-CoV-2 variants with meaningful reasons.

MFTrans: A multi-feature transformer network for protein secondary structure prediction.

In the rapidly evolving field of computational biology, accurate prediction of protein secondary structures is crucial for understanding protein functions, facilitating drug discovery, and advancing disease diagnostics. In this paper, we propose MFTrans, a deep learning-based multi-feature fusion network aimed at enhancing the precision and efficiency of Protein Secondary Structure Prediction (PSSP). This model employs a Multiple Sequence Alignment (MSA) Transformer in combination with a multi-view deep learning architecture to effectively capture both global and local features of protein sequences. MFTrans integrates diverse features generated by protein sequences, including MSA, sequence information, evolutionary information, and hidden state information, using a multi-feature fusion strategy. The MSA Transformer is utilized to interleave row and column attention across the input MSA, while a Transformer encoder and decoder are introduced to enhance the extracted high-level features. A hybrid network architecture, combining a convolutional neural network with a bidirectional Gated Recurrent Unit (BiGRU) network, is used to further extract high-level features after feature fusion. In independent tests, our experimental results show that MFTrans has superior generalization ability, outperforming other state-of-the-art PSSP models by 3 % on average on public benchmarks including CASP12, CASP13, CASP14, TEST2016, TEST2018, and CB513. Case studies further highlight its advanced performance in predicting mutation sites. MFTrans contributes significantly to the protein science field, opening new avenues for drug discovery, disease diagnosis, and protein.

Transmembrane determinants of voltage-gating differences between BK (Slo1) and Slo3 channels.

Voltage-gated potassium channels are critical in modulating cellular excitability, with Slo (slowpoke) channels forming a unique family characterized by their large conductance and dual regulation by electrical signals and intracellular messengers. Despite their structural and evolutionary similarities, Slo1 and Slo3 channels exhibit significant differences in their voltage-gating properties. This study investigates the molecular determinants that differentiate the voltage-gating properties of human Slo1 and mouse Slo3 channels. Utilizing Slo1/Slo3 chimeras, we pinpointed the selectivity filter region as a key factor in the Slo3 channel's reduced conductance at negative voltages. The S6 transmembrane (TM) segment was identified as pivotal for the Slo3 channel's biphasic deactivation kinetics at these voltages. Additionally, the S4 and S6 TM segments were found to be responsible for the gradual slope in the Slo3 channel's conductance-voltage relationship, while multiple TM regions appear to be involved in the Slo3 channel's requirement of strong depolarization for activation. Mutations in the Slo1's S5 and S6 TM segments revealed three residues (I233, L302, and M304) that likely play a crucial role in the allosteric coupling between the voltage sensors and the pore gate.

Dynamic network biomarker C1QTNF1 regulates tumor formation at the tipping point of hepatocellular carcinoma.

Identifying the precise moment before the onset of hepatocellular carcinoma (HCC) remains a significant challenge in the medical field. The existing biomarkers fall short of pinpointing the critical point preceding HCC formation. This study aimed to determine the exact tipping point for the transition from cirrhosis to HCC, identify the core Dynamic Network Biomarker (DNB), and elucidate its regulatory effects on HCC. A spontaneous HCC mouse model was established to mimic HCC formation in patients with chronic hepatitis. Using the DNB method, C1q and tumor necrosis factor (TNF) related 1 (C1QTNF1) protein was identified as the key DNB at the crucial tipping time of spontaneous HCC development. Both in vitro and in vivo studies showed that C1QTNF1 could inhibit tumor growth. Overexpression of C1QTNF1 before the tipping point effectively prevented HCC occurrence. Patients with elevated C1QTNF1 expression demonstrated improved overall survival (OS) (P = 0.03) and disease-free survival (DFS) (P = 0.03). The diagnostic value of C1QTNF1 was comparable to that of alpha-fetoprotein (AFP) (area under the curve [AUC] = 0.84; sensitivity 85%; specificity 80%). Furthermore, our research indicated that platelet-expressed C1QTNF1 is involved in cancer-associated signaling pathways. Our findings introduce a novel perspective by highlighting C1QTNF1 as the pivotal biomarker at the tipping point of primary HCC formation using DNB. We propose C1QTNF1 as a prognostic biomarker for HCC, potentially influencing tumor development through a platelet-related cancer signaling pathway.

PSSP-MFFNet: A Multifeature Fusion Network for Protein Secondary Structure Prediction.

Protein secondary structure prediction (PSSP) is a fundamental task in modern bioinformatics research and is particularly important for uncovering the functional mechanisms of proteins. To improve the accuracy of PSSP, various general and essential features generated from amino acid sequences are often used for predicting the secondary structure. In this paper, we propose PSSP-MFFNet, a deep learning-based multi-feature fusion network for PSSP, which incorporates a multi-view deep learning architecture with the multiple sequence alignment (MSA) Transformer to efficiently capture global and local features of protein sequences. In practice, PSSP-MFFNet adopts a multi-feature fusion strategy, integrating different features generated from protein sequences, including MSA, sequence information, evolutionary information, and hidden state information. Moreover, we employ the MSA Transformer to interleave row and column attention across the input MSA. A hybrid network architecture of convolutional neural networks and long short-term memory networks is applied to extract high-level features after feature fusion. Furthermore, we introduce a transformer encoder to enhance the extracted high-level features. Comparative experimental results on independent tests demonstrate that PSSP-MFFNet has excellent generalization ability, outperforming other state-of-the-art PSSP models by an average of 1% on public benchmarks, including CASP12, CASP13, CASP14, TEST2018, and CB513. Our method can contribute to a better understanding of the biological functions of proteins, which has significant implications for drug discovery, disease diagnosis, and protein engineering.

MRI-based automated machine learning model for preoperative identification of variant histology in muscle-invasive bladder carcinoma.

OBJECTIVES: It is essential yet highly challenging to preoperatively diagnose variant histologies such as urothelial carcinoma with squamous differentiation (UC w/SD) from pure UC in patients with muscle-invasive bladder carcinoma (MIBC), as their treatment strategy varies significantly. We developed a non-invasive automated machine learning (AutoML) model to preoperatively differentiate UC w/SD from pure UC in patients with MIBC. METHODS: A total of 119 MIBC patients who underwent baseline bladder MRI were enrolled in this study, including 38 patients with UC w/SD and 81 patients with pure UC. These patients were randomly assigned to a training set or a test set (3:1). An AutoML model was built from the training set, using 13 selected radiomic features from T2-weighted imaging, semantic features (ADC values), and clinical features (tumor length, tumor stage, lymph node metastasis status), and subsequent ten-fold cross-validation was performed. A test set was used to validate the proposed model. The AUC of the ROC curve was then calculated for the model. RESULTS: This AutoML model enabled robust differentiation of UC w/SD and pure UC in patients with MIBC in both training set (ten-fold cross-validation AUC = 0.955, 95% confidence interval [CI]: 0.944-0.965) and test set (AUC = 0.932, 95% CI: 0.812-1.000). CONCLUSION: The presented AutoML model, that incorporates the radiomic, semantic, and clinical features from baseline MRI, could be useful for preoperative differentiation of UC w/SD and pure UC. CLINICAL RELEVANCE STATEMENT: This MRI-based automated machine learning (AutoML) study provides a non-invasive and low-cost preoperative prediction tool to identify the muscle-invasive bladder cancer patients with variant histology, which may serve as a useful tool for clinical decision-making. KEY POINTS: • It is important to preoperatively diagnose variant histology from urothelial carcinoma in patients with muscle-invasive bladder carcinoma (MIBC), as their treatment strategy varies significantly. • An automated machine learning (AutoML) model based on baseline bladder MRI can identify the variant histology (squamous differentiation) from urothelial carcinoma preoperatively in patients with MIBC. • The developed AutoML model is a non-invasive and low-cost preoperative prediction tool, which may be useful for clinical decision-making.

Discovery of Kinetin in inhibiting colorectal cancer progression via enhancing PSMB1-mediated RAB34 degradation.

Colorectal cancer (CRC) is one of the most prevalent malignancies worldwide. Understanding the underlying mechanism driving CRC progression and identifying potential therapeutic drug targets are of utmost urgency. We previously utilized LC-MS-based proteomic profiling to identify proteins associated with postoperative progression in stage II/III CRC. Here, we revealed that proteasome subunit beta type-1 (PSMB1) is an independent predictor for postoperative progression in stage II/III CRC. Mechanistically, PSMB1 binds directly to onco-protein RAB34 and promotes its proteasome-dependent degradation, potentially leading to the inactivation of the MEK/ERK signaling pathway and inhibition of CRC progression. To further identify potential anticancer drugs, we screened a library of 2509 FDA-approved drugs using computer-aided drug design (CADD) and identified Kinetin as a potentiating agent for PSMB1. Functional assays confirmed that Kinetin enhanced the interaction between PSMB1 and RAB34, hence facilitated the degradation of RAB34 protein and decreased the MEK/ERK phosphorylation. Kinetin suppresses CRC progression in patient-derived xenograft (PDX) and liver metastasis models. Conclusively, our study identifies PSMB1 as a potential biomarker and therapeutic target for CRC, and Kinetin as an anticancer drug by enhancing proteasome-dependent onco-protein degradation.

TCMBank: bridges between the largest herbal medicines, chemical ingredients, target proteins, and associated diseases with intelligence text mining.

Traditional Chinese Medicine (TCM) has long been viewed as a precious source of modern drug discovery. AI-assisted drug discovery (AIDD) has been investigated extensively. However, there are still two challenges in applying AIDD to guide TCM drug discovery: the lack of a large amount of standardized TCM-related information and AIDD is prone to pathological failures in out-of-domain data. We have released TCM Database@Taiwan in 2011, and it has been widely disseminated and used. Now, we developed TCMBank, the largest systematic free TCM database, which is an extension of TCM Database@Taiwan. TCMBank contains 9192 herbs, 61 966 ingredients (unduplicated), 15 179 targets, 32 529 diseases, and their pairwise relationships. By integrating multiple data sources, TCMBank provides 3D structure information of ingredients and provides a standard list and detailed information on herbs, ingredients, targets and diseases. TCMBank has an intelligent document identification module that continuously adds TCM-related information retrieved from the literature in PubChem. In addition, driven by TCMBank big data, we developed an ensemble learning-based drug discovery protocol for identifying potential leads and drug repurposing. We take colorectal cancer and Alzheimer's disease as examples to demonstrate how to accelerate drug discovery by artificial intelligence. Using TCMBank, researchers can view literature-driven relationship mapping between herbs/ingredients and genes/diseases, allowing the understanding of molecular action mechanisms for ingredients and identification of new potentially effective treatments. TCMBank is available at https://TCMBank.CN/.

Hybrid neural network approaches to predict drug-target binding affinity for drug repurposing: screening for potential leads for Alzheimer's disease.

Alzheimer's disease (AD) is a neurodegenerative disease that primarily affects elderly individuals. Recent studies have found that sigma-1 receptor (S1R) agonists can maintain endoplasmic reticulum stress homeostasis, reduce neuronal apoptosis, and enhance mitochondrial function and autophagy, making S1R a target for AD therapy. Traditional experimental methods are costly and inefficient, and rapid and accurate prediction methods need to be developed, while drug repurposing provides new ways and options for AD treatment. In this paper, we propose HNNDTA, a hybrid neural network for drug-target affinity (DTA) prediction, to facilitate drug repurposing for AD treatment. The study combines protein-protein interaction (PPI) network analysis, the HNNDTA model, and molecular docking to identify potential leads for AD. The HNNDTA model was constructed using 13 drug encoding networks and 9 target encoding networks with 2506 FDA-approved drugs as the candidate drug library for S1R and related proteins. Seven potential drugs were identified using network pharmacology and DTA prediction results of the HNNDTA model. Molecular docking simulations were further performed using the AutoDock Vina tool to screen haloperidol and bromperidol as lead compounds for AD treatment. Absorption, distribution, metabolism, excretion, and toxicity (ADMET) evaluation results indicated that both compounds had good pharmacokinetic properties and were virtually non-toxic. The study proposes a new approach to computer-aided drug design that is faster and more economical, and can improve hit rates for new drug compounds. The results of this study provide new lead compounds for AD treatment, which may be effective due to their multi-target action. HNNDTA is freely available at https://github.com/lizhj39/HNNDTA.

Aquaporin 9 is involved in CRC metastasis through DVL2-dependent Wnt/ß-catenin signaling activation.

Background: Aquaporin 9 (AQP9) is permeable to water or other small molecules, and plays an important role in various cancers. We previously found that AQP9 was related to the efficacy of chemotherapy in patients with colorectal cancer (CRC). This study aimed to identify the role and regulatory mechanism of AQP9 in CRC metastasis. Methods: The clinical significance of AQP9 was analysed by using bioinformatics and tissue microarray. Transcriptome sequencing, Dual-Luciferase Reporter Assay, Biacore, and co-immunoprecipitation were employed to demonstrate the regulatory mechanism of AQP9 in CRC. The relationship between AQP9 and CRC metastasis was verified in vitro and in vivo by using real-time cell analysis assay, high content screening, and liver metastasis models of nude mice. Results: We found that AQP9 was highly expressed in metastatic CRC. AQP9 overexpression reduced cell roundness and enhanced cell motility in CRC. We further showed that AQP9 interacted with Dishevelled 2 (DVL2) via the C-terminal SVIM motif, resulting in DVL2 stabilization and the Wnt/ß-catenin pathway activation. Additionally, we identified the E3 ligase neural precursor cell expressed developmentally downregulated 4-like (NEDD4L) as a modulator regulating the ubiquitination and degradation of AQP9. Conclusions: Collectively, our study revealed the important role of AQP9 in regulating DVL2 stabilization and Wnt/ß-catenin signaling to promote CRC metastasis. Targeting the NEDD4L-AQP9-DVL2 axis might have therapeutic usefulness in metastatic CRC treatment.

Tiam1 coordinates synaptic structural and functional plasticity underpinning the pathophysiology of neuropathic pain.

Neuropathic pain is a common, debilitating chronic pain condition caused by damage or a disease affecting the somatosensory nervous system. Understanding the pathophysiological mechanisms underlying neuropathic pain is critical for developing new therapeutic strategies to treat chronic pain effectively. Tiam1 is a Rac1 guanine nucleotide exchange factor (GEF) that promotes dendritic and synaptic growth during hippocampal development by inducing actin cytoskeletal remodeling. Here, using multiple neuropathic pain animal models, we show that Tiam1 coordinates synaptic structural and functional plasticity in the spinal dorsal horn via actin cytoskeleton reorganization and synaptic NMDAR stabilization and that these actions are essential for the initiation, transition, and maintenance of neuropathic pain. Furthermore, an antisense oligonucleotides (ASO) targeting spinal Tiam1 persistently alleviate neuropathic pain sensitivity. Our findings suggest that Tiam1-coordinated synaptic functional and structural plasticity underlies the pathophysiology of neuropathic pain and that intervention of Tiam1-mediated maladaptive synaptic plasticity has long-lasting consequences in neuropathic pain management.

NHGNN-DTA: a node-adaptive hybrid graph neural network for interpretable drug-target binding affinity prediction.

MOTIVATION: Large-scale prediction of drug-target affinity (DTA) plays an important role in drug discovery. In recent years, machine learning algorithms have made great progress in DTA prediction by utilizing sequence or structural information of both drugs and proteins. However, sequence-based algorithms ignore the structural information of molecules and proteins, while graph-based algorithms are insufficient in feature extraction and information interaction. RESULTS: In this article, we propose NHGNN-DTA, a node-adaptive hybrid neural network for interpretable DTA prediction. It can adaptively acquire feature representations of drugs and proteins and allow information to interact at the graph level, effectively combining the advantages of both sequence-based and graph-based approaches. Experimental results have shown that NHGNN-DTA achieved new state-of-the-art performance. It achieved the mean squared error (MSE) of 0.196 on the Davis dataset (below 0.2 for the first time) and 0.124 on the KIBA dataset (3% improvement). Meanwhile, in the case of cold start scenario, NHGNN-DTA proved to be more robust and more effective with unseen inputs than baseline methods. Furthermore, the multi-head self-attention mechanism endows the model with interpretability, providing new exploratory insights for drug discovery. The case study on Omicron variants of SARS-CoV-2 illustrates the efficient utilization of drug repurposing in COVID-19. AVAILABILITY AND IMPLEMENTATION: The source code and data are available at https://github.com/hehh77/NHGNN-DTA.

BK channel modulation by positively charged peptides and auxiliary γ subunits mediated by the Ca2+-bowl site.

The large-conductance, Ca2+-, and voltage-activated K+ (BK) channel consists of the pore-forming α (BKα) subunit and regulatory ß and γ subunits. The γ1-3 subunits facilitate BK channel activation by shifting the voltage-dependence of channel activation toward the hyperpolarization direction by about 50-150 mV in the absence of Ca2+. We previously found that the intracellular C-terminal positively charged regions of the γ subunits play important roles in BK channel modulation. In this study, we found that the intracellular C-terminal region of BKα is indispensable in BK channel modulation by the γ1 subunit. Notably, synthetic peptide mimics of the γ1-3 subunits' C-terminal positively charged regions caused 30-50 mV shifts in BKα channel voltage-gating toward the hyperpolarization direction. The cationic cell-penetrating HIV-1 Tat peptide exerted a similar BK channel-activating effect. The BK channel-activating effects of the synthetic peptides were reduced in the presence of Ca2+ and markedly ablated by both charge neutralization of the Ca2+-bowl site and high ionic strength, suggesting the involvement of electrostatic interactions. The efficacy of the γ subunits in BK channel modulation was reduced by charge neutralization of the Ca2+-bowl site. However, BK channel modulation by the γ1 subunit was little affected by high ionic strength and the positively charged peptide remained effective in BK channel modulation in the presence of the γ1 subunit. These findings identify positively charged peptides as BK channel modulators and reveal a role for the Ca2+-bowl site in BK channel modulation by positively charged peptides and the C-terminal positively charged regions of auxiliary γ subunits.

Meta Learning With Graph Attention Networks for Low-Data Drug Discovery.

Finding candidate molecules with favorable pharmacological activity, low toxicity, and proper pharmacokinetic properties is an important task in drug discovery. Deep neural networks have made impressive progress in accelerating and improving drug discovery. However, these techniques rely on a large amount of label data to form accurate predictions of molecular properties. At each stage of the drug discovery pipeline, usually, only a few biological data of candidate molecules and derivatives are available, indicating that the application of deep neural networks for low-data drug discovery is still a formidable challenge. Here, we propose a meta learning architecture with graph attention network, Meta-GAT, to predict molecular properties in low-data drug discovery. The GAT captures the local effects of atomic groups at the atom level through the triple attentional mechanism and implicitly captures the interactions between different atomic groups at the molecular level. GAT is used to perceive molecular chemical environment and connectivity, thereby effectively reducing sample complexity. Meta-GAT further develops a meta learning strategy based on bilevel optimization, which transfers meta knowledge from other attribute prediction tasks to low-data target tasks. In summary, our work demonstrates how meta learning can reduce the amount of data required to make meaningful predictions of molecules in low-data scenarios. Meta learning is likely to become the new learning paradigm in low-data drug discovery. The source code is publicly available at: https://github.com/lol88/Meta-GAT.

DSIL-DDI: A Domain-Invariant Substructure Interaction Learning for Generalizable Drug-Drug Interaction Prediction.

Drug-drug interactions (DDIs) trigger unexpected pharmacological effects in vivo, often with unknown causal mechanisms. Deep learning methods have been developed to better understand DDI. However, learning domain-invariant representations for DDI remains a challenge. Generalizable DDI predictions are closer to reality than source domain predictions. For existing methods, it is difficult to achieve out-of-distribution (OOD) predictions. In this article, focusing on substructure interaction, we propose DSIL-DDI, a pluggable substructure interaction module that can learn domain-invariant representations of DDIs from source domain. We evaluate DSIL-DDI on three scenarios: the transductive setting (all drugs in test set appear in training set), the inductive setting (test set contains new drugs that were not present in training set), and OOD generalization setting (training set and test set belong to two different datasets). The results demonstrate that DSIL-DDI improve the generalization and interpretability of DDI prediction modeling and provides valuable insights for OOD DDI predictions. DSIL-DDI can help doctors ensuring the safety of drug administration and reducing the harm caused by drug abuse.

3DGT-DDI: 3D graph and text based neural network for drug-drug interaction prediction.

MOTIVATION: Drug-drug interactions (DDIs) occur during the combination of drugs. Identifying potential DDI helps us to study the mechanism behind the combination medication or adverse reactions so as to avoid the side effects. Although many artificial intelligence methods predict and mine potential DDI, they ignore the 3D structure information of drug molecules and do not fully consider the contribution of molecular substructure in DDI. RESULTS: We proposed a new deep learning architecture, 3DGT-DDI, a model composed of a 3D graph neural network and pre-trained text attention mechanism. We used 3D molecular graph structure and position information to enhance the prediction ability of the model for DDI, which enabled us to deeply explore the effect of drug substructure on DDI relationship. The results showed that 3DGT-DDI outperforms other state-of-the-art baselines. It achieved an 84.48% macro F1 score in the DDIExtraction 2013 shared task dataset. Also, our 3D graph model proves its performance and explainability through weight visualization on the DrugBank dataset. 3DGT-DDI can help us better understand and identify potential DDI, thereby helping to avoid the side effects of drug mixing. AVAILABILITY: The source code and data are available at https://github.com/hehh77/3DGT-DDI.

The leucine-rich repeat domains of BK channel auxiliary γ subunits regulate their expression, trafficking, and channel-modulation functions.

As high-conductance calcium- and voltage-dependent potassium channels, BK channels consist of pore-forming, voltage-, and Ca2+-sensing α and auxiliary subunits. The leucine-rich repeat (LRR) domain-containing auxiliary γ subunits potently modulate the voltage dependence of BK channel activation. Despite their dominant size in whole protein masses, the function of the LRR domain in BK channel γ subunits is unknown. We here investigated the function of these LRR domains in BK channel modulation by the auxiliary γ1-3 (LRRC26, LRRC52, and LRRC55) subunits. Using cell surface protein immunoprecipitation, we validated the predicted extracellular localization of the LRR domains. We then refined the structural models of mature proteins on the membrane via molecular dynamic simulations. By replacement of the LRR domain with extracellular regions or domains of non-LRR proteins, we found that the LRR domain is nonessential for the maximal channel-gating modulatory effect but is necessary for the all-or-none phenomenon of BK channel modulation by the γ1 subunit. Mutational and enzymatic blockade of N-glycosylation in the γ1-3 subunits resulted in a reduction or loss of BK channel modulation by γ subunits. Finally, by analyzing their expression in whole cells and on the plasma membrane, we found that blockade of N-glycosylation drastically reduced total expression of the γ2 subunit and the cell surface expression of the γ1 and γ3 subunits. We conclude that the LRR domains play key roles in the regulation of the expression, cell surface trafficking, and channel-modulation functions of the BK channel γ subunits.

FusionDTA: attention-based feature polymerizer and knowledge distillation for drug-target binding affinity prediction.

The prediction of drug-target affinity (DTA) plays an increasingly important role in drug discovery. Nowadays, lots of prediction methods focus on feature encoding of drugs and proteins, but ignore the importance of feature aggregation. However, the increasingly complex encoder networks lead to the loss of implicit information and excessive model size. To this end, we propose a deep-learning-based approach namely FusionDTA. For the loss of implicit information, a novel muti-head linear attention mechanism was utilized to replace the rough pooling method. This allows FusionDTA aggregates global information based on attention weights, instead of selecting the largest one as max-pooling does. To solve the redundancy issue of parameters, we applied knowledge distillation in FusionDTA by transfering learnable information from teacher model to student. Results show that FusionDTA performs better than existing models for the test domain on all evaluation metrics. We obtained concordance index (CI) index of 0.913 and 0.906 in Davis and KIBA dataset respectively, compared with 0.893 and 0.891 of previous state-of-art model. Under the cold-start constrain, our model proved to be more robust and more effective with unseen inputs than baseline methods. In addition, the knowledge distillation did save half of the parameters of the model, with only 0.006 reduction in CI index. Even FusionDTA with half the parameters could easily exceed the baseline on all metrics. In general, our model has superior performance and improves the effect of drug-target interaction (DTI) prediction. The visualization of DTI can effectively help predict the binding region of proteins during structure-based drug design.

Mol2Context-vec: learning molecular representation from context awareness for drug discovery.

With the rapid development of proteomics and the rapid increase of target molecules for drug action, computer-aided drug design (CADD) has become a basic task in drug discovery. One of the key challenges in CADD is molecular representation. High-quality molecular expression with chemical intuition helps to promote many boundary problems of drug discovery. At present, molecular representation still faces several urgent problems, such as the polysemy of substructures and unsmooth information flow between atomic groups. In this research, we propose a deep contextualized Bi-LSTM architecture, Mol2Context-vec, which can integrate different levels of internal states to bring dynamic representations of molecular substructures. And the obtained molecular context representation can capture the interactions between any atomic groups, especially a pair of atomic groups that are topologically distant. Experiments show that Mol2Context-vec achieves state-of-the-art performance on multiple benchmark datasets. In addition, the visual interpretation of Mol2Context-vec is very close to the structural properties of chemical molecules as understood by humans. These advantages indicate that Mol2Context-vec can be used as a reliable and effective tool for molecular expression. Availability: The source code is available for download in https://github.com/lol88/Mol2Context-vec.