Improving drug response prediction via integrating gene relationships with deep learning.

Predicting the drug response of cancer cell lines is crucial for advancing personalized cancer treatment, yet remains challenging due to tumor heterogeneity and individual diversity. In this study, we present a deep learning-based framework named Deep neural network Integrating Prior Knowledge (DIPK) (DIPK), which adopts self-supervised techniques to integrate multiple valuable information, including gene interaction relationships, gene expression profiles and molecular topologies, to enhance prediction accuracy and robustness. We demonstrated the superior performance of DIPK compared to existing methods on both known and novel cells and drugs, underscoring the importance of gene interaction relationships in drug response prediction. In addition, DIPK extends its applicability to single-cell RNA sequencing data, showcasing its capability for single-cell-level response prediction and cell identification. Further, we assess the applicability of DIPK on clinical data. DIPK accurately predicted a higher response to paclitaxel in the pathological complete response (pCR) group compared to the residual disease group, affirming the better response of the pCR group to the chemotherapy compound. We believe that the integration of DIPK into clinical decision-making processes has the potential to enhance individualized treatment strategies for cancer patients.

Effects of levosimendan on the outcome of veno-arterial extracorporeal membrane oxygenation: a systematic review and meta-analysis.

OBJECTIVES: For patients with severe cardiopulmonary failure, such as cardiogenic shock, veno-arterial extracorporeal membrane oxygenation (VA-ECMO) is primarily utilized to preserve their life by providing continuous extracorporeal respiration and circulation. However, because of the complexity of patients' underlying diseases and serious complications, successful weaning from ECMO is often difficult. At present, there have been limited studies on ECMO weaning strategies, so the principal purpose of this meta-analysis is to examine how levosimendan contributes to the weaning of extracorporeal membrane oxygenation. METHODS: The Cochrane Library, Embase, Web of Science, and PubMed were browsed for all potentially related research about clinical benefits of levosimendan in weaning patients receiving VA-ECMO and included 15 of them. The main outcome is success of weaning from extracorporeal membrane oxygenation, with the secondary outcomes of 1-month mortality (28 or 30 days), ECMO duration, hospital or intensive care unit (ICU) length of stay, and use of vasoactive drugs. RESULTS: 1772 patients altogether from 15 publications were incorporated in our meta-analysis. We used fixed and random-effect models to combine odds ratio (OR) and 95% confidence interval (CI) for dichotomous outcomes and standardized mean difference (SMD) for continuous outcomes. The weaning success rate in the levosimendan group was considerably higher in contrast to the comparison (OR = 2.78, 95% CI 1.80-4.30; P < 0.00001; I2 = 65%), and subgroup analysis showed that there was less heterogeneity in patients after cardiac surgery (OR = 2.06, 95% CI, 1.35-3.12; P = 0.0007; I2 = 17%). In addition, the effect of levosimendan on improving weaning success rate was statistically significant only at 0.2 mcg/kg/min (OR = 2.45, 95% CI, 1.11-5.40; P = 0.03; I2 = 38%). At the same time, the 28-day or 30-day proportion of deaths in the sample receiving levosimendan also decreased (OR = 0.47, 95% CI, 0.28-0.79; P = 0.004; I2 = 73%), and the difference was statistically significant. In terms of secondary outcomes, we found that individuals undergoing levosimendan treatment had a longer duration of VA-ECMO support. CONCLUSIONS: In patients receiving VA-ECMO, levosimendan treatment considerably raised the weaning success rate and helped lower mortality. Since most of the evidence comes from retrospective studies, more randomized multicenter trials are required to verify the conclusion.

Improving drug-target affinity prediction via feature fusion and knowledge distillation.

Rapid and accurate prediction of drug-target affinity can accelerate and improve the drug discovery process. Recent studies show that deep learning models may have the potential to provide fast and accurate drug-target affinity prediction. However, the existing deep learning models still have their own disadvantages that make it difficult to complete the task satisfactorily. Complex-based models rely heavily on the time-consuming docking process, and complex-free models lacks interpretability. In this study, we introduced a novel knowledge-distillation insights drug-target affinity prediction model with feature fusion inputs to make fast, accurate and explainable predictions. We benchmarked the model on public affinity prediction and virtual screening dataset. The results show that it outperformed previous state-of-the-art models and achieved comparable performance to previous complex-based models. Finally, we study the interpretability of this model through visualization and find it can provide meaningful explanations for pairwise interaction. We believe this model can further improve the drug-target affinity prediction for its higher accuracy and reliable interpretability.

Molecular language models: RNNs or transformer?

Language models have shown the capacity to learn complex molecular distributions. In the field of molecular generation, they are designed to explore the distribution of molecules, and previous studies have demonstrated their ability to learn molecule sequences. In the early times, recurrent neural networks (RNNs) were widely used for feature extraction from sequence data and have been used for various molecule generation tasks. In recent years, the attention mechanism for sequence data has become popular. It captures the underlying relationships between words and is widely applied to language models. The Transformer-Layer, a model based on a self-attentive mechanism, also shines the same as the RNN-based model. In this research, we investigated the difference between RNNs and the Transformer-Layer to learn a more complex distribution of molecules. For this purpose, we experimented with three different generative tasks: the distributions of molecules with elevated scores of penalized LogP, multimodal distributions of molecules and the largest molecules in PubChem. We evaluated the models on molecular properties, basic metrics, Tanimoto similarity, etc. In addition, we applied two different representations of the molecule, SMILES and SELFIES. The results show that the two language models can learn complex molecular distributions and SMILES-based representation has better performance than SELFIES. The choice between RNNs and the Transformer-Layer needs to be based on the characteristics of dataset. RNNs work better on data focus on local features and decreases with multidistribution data, while the Transformer-Layer is more suitable when meeting molecular with larger weights and focusing on global features.

Deep generative model for drug design from protein target sequence.

Drug discovery for a protein target is a laborious and costly process. Deep learning (DL) methods have been applied to drug discovery and successfully generated novel molecular structures, and they can substantially reduce development time and costs. However, most of them rely on prior knowledge, either by drawing on the structure and properties of known molecules to generate similar candidate molecules or extracting information on the binding sites of protein pockets to obtain molecules that can bind to them. In this paper, DeepTarget, an end-to-end DL model, was proposed to generate novel molecules solely relying on the amino acid sequence of the target protein to reduce the heavy reliance on prior knowledge. DeepTarget includes three modules: Amino Acid Sequence Embedding (AASE), Structural Feature Inference (SFI), and Molecule Generation (MG). AASE generates embeddings from the amino acid sequence of the target protein. SFI inferences the potential structural features of the synthesized molecule, and MG seeks to construct the eventual molecule. The validity of the generated molecules was demonstrated by a benchmark platform of molecular generation models. The interaction between the generated molecules and the target proteins was also verified on the basis of two metrics, drug-target affinity and molecular docking. The results of the experiments indicated the efficacy of the model for direct molecule generation solely conditioned on amino acid sequence.

Development, validation, and evaluation of a deep learning model to screen cyclin-dependent kinase 12 inhibitors in cancers.

Deep learning-based in silico alternatives have been demonstrated to be of significant importance in the acceleration of the drug discovery process and enhancement of success rates. Cyclin-dependent kinase 12 (CDK12) is a transcription-related cyclin-dependent kinase that may act as a biomarker and therapeutic target for cancers. However, currently, there is no high selective CDK12 inhibitor in clinical development and the identification of new specific CDK12 inhibitors has become increasingly challenging due to their similarity with CDK13. In this study, we developed a virtual screening workflow that combines deep learning with virtual screening tools and can be applied rapidly to millions of molecules. We designed a Transformer architecture Drug-Target Interaction (DTI) model with dual-branched self-supervised pre-trained molecular graph models and protein sequence models. Our predictive model produced satisfactory predictions for various targets, including CDK12, with several novel hits. We screened a large compound library consisting of 4.5 million drug-like molecules and recommended a list of potential CDK12 inhibitors for further experimental testing. In kinase assay, compared to the positive CDK12 inhibitor THZ531, the compounds CICAMPA-01, 02, 03 displayed more effective inhibition of CDK12, up to three times as much as THZ531. The compounds CICAMPA-03, 05, 04, 07 showed less inhibition of CDK13 compare to THZ531. In vitro, the IC50 of CICAMPA-01, 04, 05, 06, 09 was less than 3 µM in the HER2 positive CDK12 amplification breast cancer cell line BT-474. Overall, this study provides a highly efficient and end-to-end deep learning protocol, in conjunction with molecular docking, for discovering CDK12 inhibitors in cancers. Additionally, we disclose five novel CDK12 inhibitors. These results may accelerate the discovery of novel chemical-class drugs for cancer treatment.

The effect of the cyclic GMP-AMP synthase-stimulator of interferon genes signaling pathway on organ inflammatory injury and fibrosis.

The cyclic GMP-AMP synthase-stimulator of interferon genes signal transduction pathway is critical in innate immunity, infection, and inflammation. In response to pathogenic microbial infections and other conditions, cyclic GMP-AMP synthase (cGAS) recognizes abnormal DNA and initiates a downstream type I interferon response. This paper reviews the pathogenic mechanisms of stimulator of interferon genes (STING) in different organs, including changes in fibrosis-related biomarkers, intending to systematically investigate the effect of the cyclic GMP-AMP synthase-stimulator of interferon genes signal transduction in inflammation and fibrosis processes. The effects of stimulator of interferon genes in related auto-inflammatory and neurodegenerative diseases are described in this article, in addition to the application of stimulator of interferon genes-related drugs in treating fibrosis.

Deep geometric representations for modeling effects of mutations on protein-protein binding affinity.

Modeling the impact of amino acid mutations on protein-protein interaction plays a crucial role in protein engineering and drug design. In this study, we develop GeoPPI, a novel structure-based deep-learning framework to predict the change of binding affinity upon mutations. Based on the three-dimensional structure of a protein, GeoPPI first learns a geometric representation that encodes topology features of the protein structure via a self-supervised learning scheme. These representations are then used as features for training gradient-boosting trees to predict the changes of protein-protein binding affinity upon mutations. We find that GeoPPI is able to learn meaningful features that characterize interactions between atoms in protein structures. In addition, through extensive experiments, we show that GeoPPI achieves new state-of-the-art performance in predicting the binding affinity changes upon both single- and multi-point mutations on six benchmark datasets. Moreover, we show that GeoPPI can accurately estimate the difference of binding affinities between a few recently identified SARS-CoV-2 antibodies and the receptor-binding domain (RBD) of the S protein. These results demonstrate the potential of GeoPPI as a powerful and useful computational tool in protein design and engineering. Our code and datasets are available at: https://github.com/Liuxg16/GeoPPI.

An effective self-supervised framework for learning expressive molecular global representations to drug discovery.

How to produce expressive molecular representations is a fundamental challenge in artificial intelligence-driven drug discovery. Graph neural network (GNN) has emerged as a powerful technique for modeling molecular data. However, previous supervised approaches usually suffer from the scarcity of labeled data and poor generalization capability. Here, we propose a novel molecular pre-training graph-based deep learning framework, named MPG, that learns molecular representations from large-scale unlabeled molecules. In MPG, we proposed a powerful GNN for modelling molecular graph named MolGNet, and designed an effective self-supervised strategy for pre-training the model at both the node and graph-level. After pre-training on 11 million unlabeled molecules, we revealed that MolGNet can capture valuable chemical insights to produce interpretable representation. The pre-trained MolGNet can be fine-tuned with just one additional output layer to create state-of-the-art models for a wide range of drug discovery tasks, including molecular properties prediction, drug-drug interaction and drug-target interaction, on 14 benchmark datasets. The pre-trained MolGNet in MPG has the potential to become an advanced molecular encoder in the drug discovery pipeline.

TrimNet: learning molecular representation from triplet messages for biomedicine.

MOTIVATION: Computational methods accelerate drug discovery and play an important role in biomedicine, such as molecular property prediction and compound-protein interaction (CPI) identification. A key challenge is to learn useful molecular representation. In the early years, molecular properties are mainly calculated by quantum mechanics or predicted by traditional machine learning methods, which requires expert knowledge and is often labor-intensive. Nowadays, graph neural networks have received significant attention because of the powerful ability to learn representation from graph data. Nevertheless, current graph-based methods have some limitations that need to be addressed, such as large-scale parameters and insufficient bond information extraction. RESULTS: In this study, we proposed a graph-based approach and employed a novel triplet message mechanism to learn molecular representation efficiently, named triplet message networks (TrimNet). We show that TrimNet can accurately complete multiple molecular representation learning tasks with significant parameter reduction, including the quantum properties, bioactivity, physiology and CPI prediction. In the experiments, TrimNet outperforms the previous state-of-the-art method by a significant margin on various datasets. Besides the few parameters and high prediction accuracy, TrimNet could focus on the atoms essential to the target properties, providing a clear interpretation of the prediction tasks. These advantages have established TrimNet as a powerful and useful computational tool in solving the challenging problem of molecular representation learning. AVAILABILITY: The quantum and drug datasets are available on the website of MoleculeNet: http://moleculenet.ai. The source code is available in GitHub: https://github.com/yvquanli/trimnet. CONTACT: xjyao@lzu.edu.cn, songsen@tsinghua.edu.cn.

In Situ Growth of a Cationic Polymer from the N-Terminus of Glucose Oxidase To Regulate H2O2 Generation for Cancer Starvation and H2O2 Therapy.

Hydrogen peroxide (H2O2)-generating enzymes (HGEs) are potentially useful for tumor therapy, but the potential is limited by the challenge in regulating H2O2 production. Herein, we present site-specific in situ growth of a cationic polymer poly( N, N'-dimethylamino-2-ethyl methacrylate) (PDMA) from the N-terminus of glucose oxidase (GOX) to generate a site-specific and cationic GOX-PDMA conjugate with well-retained activity and enhanced stability to regulate H2O2 generation for cancer starvation and H2O2 therapy. Notably, the efficiency of endocytosis of the conjugate was 4-fold higher than that of free GOX. As a result, relative to free GOX, the conjugate showed 1.5-fold increased cytotoxicity, 2-fold enhanced tumor retention, and 5-fold increased tolerability after intratumoral injection. Importantly, a single intratumoral injection of the conjugate completely abolished colon tumors without detectable side effects, whereas free GOX was ineffective and systemically toxic. This chemistry may provide a new, simple, general, and efficient solution to regulate H2O2 production and thereby to dramatically improve the antitumor efficacy of HGEs while reducing side effects.

Site-Selective in Situ Growth-Induced Self-Assembly of Protein-Polymer Conjugates into pH-Responsive Micelles for Tumor Microenvironment Triggered Fluorescence Imaging.

Self-assembly of site-selective protein-polymer conjugates into stimuli-responsive micelles is interesting owing to their potential biomedical applications, ranging from molecular imaging to drug delivery, but remains a significant challenge. Herein we report a method of site-selective in situ growth-induced self-assembly (SIGS) to synthesize site-specific human serum albumin-poly(2-(diisopropylamino)ethyl methacrylate) (HSA-PDPA) conjugates that can in situ self-assemble into pH-responsive micelles with tunable morphologies. Indocyanine green (ICG) was selectively loaded into the core of sphere-like HSA-PDPA micelles to form pH-responsive fluorescence nanoprobes. The nanoprobes rapidly dissociated into protonated individual unimers at a transition pH of around 6.5, that is the extracellular pH of tumors, which resulted in a sharp fluorescence increase and markedly enhanced cellular uptake. In a tumor-bearing mouse model, they exhibited greatly enhanced tumor fluorescence imaging as compared to ICG alone and pH-nonresponsive nanoprobes. These findings suggest that pH-responsive and site-selective protein-polymer conjugate micelles synthesized by SIGS are promising as a new class of tumor microenvironment-responsive nanocarriers for enhanced tumor imaging and therapy.