A systematic evaluation of deep learning methods for the prediction of drug synergy in cancer

MMSyn: A New Multimodal Deep Learning Framework for Enhanced Prediction of Synergistic Drug Combinations

Combination therapy is a promising strategy for the successful treatment of cancer. The large number of possible combinations, however, mean that it is laborious and expensive to screen for synergistic drug combinations in vitro. Nevertheless, because of the availability of high-throughput screening data and advances in computational techniques, deep learning (DL) can be a useful tool for the prediction of synergistic drug combinations. In this study, we proposed a multimodal DL framework, MMSyn, for the prediction of synergistic drug combinations. First, features embedded in the drug molecules were extracted: structure, fingerprint, and string encoding. Then, gene expression data, DNA copy number, and pathway activity were used to describe cancer cell lines. Finally, these processed features were integrated using an attention mechanism and an interaction module and then input into a multilayer perceptron to predict drug synergy. Experimental results showed that our method outperformed five state-of-the-art DL methods and three traditional machine learning models for drug combination prediction. We verified that MMSyn achieved superior performance in stratified cross-validation settings using both the drug combination and cell line data. Moreover, we performed a set of ablation experiments to illustrate the effectiveness of each component and the efficacy of our model. In addition, our visual representation and case studies further confirmed the effectiveness of our model. All results showed that MMSyn can be used as a powerful tool for the prediction of synergistic drug combinations.

Up to the Herculean Task of Tackling Cancer Therapy Resistance

Cancer therapy resistance still poses the biggest hurdle to cancer treatment [...].

Optimized models and deep learning methods for drug response prediction in cancer treatments: a review

Recent advancements in deep learning (DL) have played a crucial role in aiding experts to develop personalized healthcare services, particularly in drug response prediction (DRP) for cancer patients. The DL's techniques contribution to this field is significant, and they have proven indispensable in the medical field. This review aims to analyze the diverse effectiveness of various DL models in making these predictions, drawing on research published from 2017 to 2023. We utilized the VOS-Viewer 1.6.18 software to create a word cloud from the titles and abstracts of the selected studies. This study offers insights into the focus areas within DL models used for drug response. The word cloud revealed a strong link between certain keywords and grouped themes, highlighting terms such as deep learning, machine learning, precision medicine, precision oncology, drug response prediction, and personalized medicine. In order to achieve an advance in DRP using DL, the researchers need to work on enhancing the models' generalizability and interoperability. It is also crucial to develop models that not only accurately represent various architectures but also simplify these architectures, balancing the complexity with the predictive capabilities. In the future, researchers should try to combine methods that make DL models easier to understand; this will make DRP reviews more open and help doctors trust the decisions made by DL models in cancer DRP.

A review of mechanistic learning in mathematical oncology

Mechanistic learning refers to the synergistic combination of mechanistic mathematical modeling and data-driven machine or deep learning. This emerging field finds increasing applications in (mathematical) oncology. This review aims to capture the current state of the field and provides a perspective on how mechanistic learning may progress in the oncology domain. We highlight the synergistic potential of mechanistic learning and point out similarities and differences between purely data-driven and mechanistic approaches concerning model complexity, data requirements, outputs generated, and interpretability of the algorithms and their results. Four categories of mechanistic learning (sequential, parallel, extrinsic, intrinsic) of mechanistic learning are presented with specific examples. We discuss a range of techniques including physics-informed neural networks, surrogate model learning, and digital twins. Example applications address complex problems predominantly from the domain of oncology research such as longitudinal tumor response predictions or time-to-event modeling. As the field of mechanistic learning advances, we aim for this review and proposed categorization framework to foster additional collaboration between the data- and knowledge-driven modeling fields. Further collaboration will help address difficult issues in oncology such as limited data availability, requirements of model transparency, and complex input data which are embraced in a mechanistic learning framework.

Heterogeneous Expression Patterns of the Minichromosome Maintenance Complex Members in Retinoblastoma Unveil Its Clinical Significance

Purpose

To explore the expression patterns and clinical significance of minichromosome maintenance (MCM) complex members in retinoblastoma (RB).

Methods

Single-cell RNA sequencing datasets from five normal retina, six intraocular, and five extraocular RB samples were integrated to characterize the expression patterns of MCM complex members at the single-cell level. Western blot and quantitative PCR were used to detect the expression of MCM complex members in RB cell lines. Immunohistochemistry was conducted to validate the expression of MCM complex members in RB patient samples and a RB mouse model.

Results

The expression of MCM2-7 is increased in RB tissue, with MCM2/3/7 showing particularly higher levels in extraocular RB. MCM3/7 are abundantly detected in cell types associated with oncogenesis. Both mRNA and protein levels of MCM3/4/6/7 are increased in RB cell lines. Immunohistochemistry further confirmed the elevated expression of MCM3 in extraocular RB, with MCM6 being the most abundantly expressed MCM in RB.

Conclusions

The distinct MCM expression patterns across various RB cell types suggest diverse functional roles, offering valuable insights for targeted therapeutic strategies. The upregulation of MCM3, MCM4, MCM6, and MCM7 in RB, with a specific emphasis on MCM6 as a notable marker, highlights their potential significance.

Mitophagy-promoting agents and their ability to promote healthy-aging

The removal of damaged mitochondrial components through a process called mitochondrial autophagy (mitophagy) is essential for the proper function of the mitochondrial network. Hence, mitophagy is vital for the health of all aerobic animals, including humans. Unfortunately, mitophagy declines with age. Many age-associated diseases, including Alzheimer's and Parkinson's, are characterized by the accumulation of damaged mitochondria and oxidative damage. Therefore, activating the mitophagy process with small molecules is an emerging strategy for treating multiple aging diseases. Recent studies have identified natural and synthetic compounds that promote mitophagy and lifespan. This article aims to summarize the existing knowledge about these substances. For readers' convenience, the knowledge is presented in a table that indicates the chemical data of each substance and its effect on lifespan. The impact on healthspan and the molecular mechanism is reported if known. The article explores the potential of utilizing a combination of mitophagy-inducing drugs within a therapeutic framework and addresses the associated challenges of this strategy. Finally, we discuss the process that balances mitophagy, i.e. mitochondrial biogenesis. In this process, new mitochondrial components are generated to replace the ones cleared by mitophagy. Furthermore, some mitophagy-inducing substances activate biogenesis (e.g. resveratrol and metformin). Finally, we discuss the possibility of combining mitophagy and biogenesis enhancers for future treatment. In conclusion, this article provides an up-to-date source of information about natural and synthetic substances that activate mitophagy and, hopefully, stimulates new hypotheses and studies that promote healthy human aging worldwide.

The recent progress of deep-learning-based in silico prediction of drug combination

Drug combination therapy has become a common strategy for the treatment of complex diseases. There is an urgent need for computational methods to efficiently identify appropriate drug combinations owing to the high cost of experimental screening. In recent years, deep learning has been widely used in the field of drug discovery. Here, we provide a comprehensive review on deep-learning-based drug combination prediction algorithms from multiple aspects. Current studies highlight the flexibility of this technology in integrating multimodal data and the ability to achieve state-of-art performance; it is expected that deep-learning-based prediction of drug combinations should play an important part in future drug discovery.