Finding melanoma drugs through a probabilistic knowledge graph

Knowledge graph applications and multi-relation learning for drug repurposing: A scoping review

OBJECTIVE

Development of novel drug solutions has always been an expensive endeavour, hence drug repurposing as an approach has gained popularity in recent years. In this review we intend to examine one of the most unique computational methods for drug repurposing, that being knowledge graphs.

METHOD

Through literature review we looked at the application of knowledge graphs in medicine, specifically at its use in drug repurposing. We also looked at literature embedding methods, integration of machine learning models and approaches to completion of knowledge graphs.

RESULT

After filtering 43 papers were used for analysis. Timeline, country distribution, application areas of knowledge graph was highlighted. General trends in the use of knowledge graphs for drug repurposing and any shortcomings of the approach was discussed.

CONCLUSION

This approach has gained popularity only very recently; hence it is in a nascent phase.

Bambu and its applications in the discovery of active molecules against melanoma

PURPOSE OR OBJECTIVE

Melanoma is one of the most dangerous forms of skin cancer and the discovery of novel drugs is an ongoing effort. Quantitative Structure Activity Relationship (QSAR) is a computational method that allows the estimation of the properties of a molecule, including its biological activity. QSAR models have been widely employed in the search for potential drug candidates, but also for agrochemicals and other molecules with applications in different branches of the industry. Here we present Bambu, a simple command line tool to generate QSAR models from high-throughput screening bioassays datasets.

METHODS

The tool was developed using the Python programming language and relies mainly on RDKit for molecule data manipulation, FLAML for automated machine learning and the PubChem REST API for data retrieval. As a proof-of-concept we have employed the tool to generate QSAR models for melanoma cell growth inhibition based on HTS data and used them to screen libraries of FDA-approved drugs and natural compounds. Additionally, Bambu was compared to QSAR-Co, another automated tool for QSAR model generation.

RESULTS

based on the developed tool we were able to produce QSAR models and identify a wide variety of molecules with potential melanoma cell growth inhibitors, many of which with anti-tumoral activity already described. The QSAR models are available through the URL http://caramel.ufpel.edu.br, and all data and code used to generate its models are available at Zenodo (https://doi.org/10.5281/zenodo.7495214). Bambu source code is available at GitHub (https://github.com/omixlab/bambu-v2). In the benchmark, Bambu was able to produce models with higher accuracy, recall, F1 and ROC AUC when compared to QSAR-Co for the selected datasets.

CONCLUSIONS

Bambu is an free and open source tool which facilitates the creation of QSAR models and can be futurely applied in a wide variety of drug discovery projects.

Drug repositioning: Progress and challenges in drug discovery for various diseases

Compared with traditional de novo drug discovery, drug repurposing has become an attractive drug discovery strategy due to its low-cost and high efficiency. Through a comprehensive analysis of the candidates that have been identified with drug repositioning potentials, it is found that although some drugs do not show obvious advantages in the original indications, they may exert more obvious effects in other diseases. In addition, some drugs have a synergistic effect to exert better clinical efficacy if used in combination. Particularly, it has been confirmed that drug repositioning has benefits and values on the current public health emergency such as the COVID-19 pandemic, which proved the great potential of drug repositioning. In this review, we systematically reviewed a series of representative drugs that have been repositioned for different diseases and illustrated successful cases in each disease. Especially, the mechanism of action for the representative drugs in new indications were explicitly explored for each disease, we hope this review can provide important insights for follow-up research.

Evaluating the performance of drug-repurposing technologies

Drug-repurposing technologies are growing in number and maturing. However, comparisons to each other and to reality are hindered because of a lack of consensus with respect to performance evaluation. Such comparability is necessary to determine scientific merit and to ensure that only meaningful predictions from repurposing technologies carry through to further validation and eventual patient use. Here, we review and compare performance evaluation measures for these technologies using version 2 of our shotgun repurposing Computational Analysis of Novel Drug Opportunities (CANDO) platform to illustrate their benefits, drawbacks, and limitations. Understanding and using different performance evaluation metrics ensures robust cross-platform comparability, enabling us to continue to strive toward optimal repurposing by decreasing the time and cost of drug discovery and development.

Toward better drug discovery with knowledge graph

Drug discovery is the process of new drug identification. This process is driven by the increasing data from existing chemical libraries and data banks. The knowledge graph is introduced to the domain of drug discovery for imposing an explicit structure to integrate heterogeneous biomedical data. The graph can provide structured relations among multiple entities and unstructured semantic relations associated with entities. In this review, we summarize knowledge graph-based works that implement drug repurposing and adverse drug reaction prediction for drug discovery. As knowledge representation learning is a common way to explore knowledge graphs for prediction problems, we introduce several representative embedding models to provide a comprehensive understanding of knowledge representation learning.

Search, access, and explore life science nanopublications on the Web

Nanopublications are Resource Description Framework (RDF) graphs encoding scientific facts extracted from the literature and enriched with provenance and attribution information. There are millions of nanopublications currently available on the Web, especially in the life science domain. Nanopublications are thought to facilitate the discovery, exploration, and re-use of scientific facts. Nevertheless, they are still not widely used by scientists outside specific circles; they are hard to find and rarely cited. We believe this is due to the lack of services to seek, find and understand nanopublications' content. To this end, we present the NanoWeb application to seamlessly search, access, explore, and re-use the nanopublications publicly available on the Web. For the time being, NanoWeb focuses on the life science domain where the vastest amount of nanopublications are available. It is a unified access point to the world of nanopublications enabling search over graph data, direct connections to evidence papers, and scientific curated databases, and visual and intuitive exploration of the relation network created by the encoded scientific facts.

Enabling Web-scale data integration in biomedicine through Linked Open Data

The biomedical data landscape is fragmented with several isolated, heterogeneous data and knowledge sources, which use varying formats, syntaxes, schemas, and entity notations, existing on the Web. Biomedical researchers face severe logistical and technical challenges to query, integrate, analyze, and visualize data from multiple diverse sources in the context of available biomedical knowledge. Semantic Web technologies and Linked Data principles may aid toward Web-scale semantic processing and data integration in biomedicine. The biomedical research community has been one of the earliest adopters of these technologies and principles to publish data and knowledge on the Web as linked graphs and ontologies, hence creating the Life Sciences Linked Open Data (LSLOD) cloud. In this paper, we provide our perspective on some opportunities proffered by the use of LSLOD to integrate biomedical data and knowledge in three domains: (1) pharmacology, (2) cancer research, and (3) infectious diseases. We will discuss some of the major challenges that hinder the wide-spread use and consumption of LSLOD by the biomedical research community. Finally, we provide a few technical solutions and insights that can address these challenges. Eventually, LSLOD can enable the development of scalable, intelligent infrastructures that support artificial intelligence methods for augmenting human intelligence to achieve better clinical outcomes for patients, to enhance the quality of biomedical research, and to improve our understanding of living systems.

Drug repositioning using drug-disease vectors based on an integrated network

BACKGROUND

Diverse interactions occur between biomolecules, such as activation, inhibition, expression, or repression. However, previous network-based studies of drug repositioning have employed interaction on the binary protein-protein interaction (PPI) network without considering the characteristics of the interactions. Recently, some studies of drug repositioning using gene expression data found that associations between drug and disease genes are useful information for identifying novel drugs to treat diseases. However, the gene expression profiles for drugs and diseases are not always available. Although gene expression profiles of drugs and diseases are available, existing methods cannot use the drugs or diseases, when differentially expressed genes in the profiles are not included in their network.

RESULTS

We developed a novel method for identifying candidate indications of existing drugs considering types of interactions between biomolecules based on known drug-disease associations. To obtain associations between drug and disease genes, we constructed a directed network using protein interaction and gene regulation data obtained from various public databases providing diverse biological pathways. The network includes three types of edges depending on relationships between biomolecules. To quantify the association between a target gene and a disease gene, we explored the shortest paths from the target gene to the disease gene and calculated the types and weights of the shortest paths. For each drug-disease pair, we built a vector consisting of values for each disease gene influenced by the drug. Using the vectors and known drug-disease associations, we constructed classifiers to identify novel drugs for each disease.

CONCLUSION

We propose a method for exploring candidate drugs of diseases using associations between drugs and disease genes derived from a directed gene network instead of gene regulation data obtained from gene expression profiles. Compared to existing methods that require information on gene relationships and gene expression data, our method can be applied to a greater number of drugs and diseases. Furthermore, to validate our predictions, we compared the predictions with drug-disease pairs in clinical trials using the hypergeometric test, which showed significant results. Our method also showed better performance compared to existing methods for the area under the receiver operating characteristic curve (AUC).

Quantitative comparison of the spreading and invasion of radial growth phase and metastatic melanoma cells in a three-dimensional human skin equivalent model

BACKGROUND

Standard two-dimensional (2D) cell migration assays do not provide information about vertical invasion processes, which are critical for melanoma progression. We provide information about three-dimensional (3D) melanoma cell migration, proliferation and invasion in a 3D melanoma skin equivalent (MSE) model. In particular, we pay careful attention to compare the structure of the tissues in the MSE with similarly-prepared 3D human skin equivalent (HSE) models. The HSE model is identically prepared to the MSE model except that melanoma cells are omitted. Using the MSE model, we examine melanoma migration, proliferation and invasion from two different human melanoma cell lines. One cell line, WM35, is associated with the early phase of the disease where spreading is thought to be confined to the epidermis. The other cell line, SK-MEL-28, is associated with the later phase of the disease where spreading into the dermis is expected.

METHODS

3D MSE and HSE models are constructed using human de-epidermised dermis (DED) prepared from skin tissue. Primary fibroblasts and primary keratinocytes are used in the MSE and HSE models to ensure the formation of a stratified epidermis, with a well-defined basement membrane. Radial spreading of cells across the surface of the HSE and MSE models is observed. Vertical invasion of melanoma cells downward through the skin is observed and measured using immunohistochemistry. All measurements of invasion are made at day 0, 9, 15 and 20, providing detailed time course data.

RESULTS

Both HSE and MSE models are similar to native skin in vivo, with a well-defined stratification of the epidermis that is separated from the dermis by a basement membrane. In the HSE and MSE we find fibroblast cells confined to the dermis, and differentiated keratinocytes in the epidermis. In the MSE, melanoma cells form colonies in the epidermis during the early part of the experiment. In the later stage of the experiment, the melanoma cells in the MSE invade deeper into the tissues. Interestingly, both the WM35 and SK-MEL-28 melanoma cells lead to a breakdown of the basement membrane and eventually enter the dermis. However, these two cell lines invade at different rates, with the SK-MEL-28 melanoma cells invading faster than the WM35 cells.

DISCUSSION

The MSE and HSE models are a reliable platform for studying melanoma invasion in a 3D tissue that is similar to native human skin. Interestingly, we find that the WM35 cell line, that is thought to be associated with radial spreading only, is able to invade into the dermis. The vertical invasion of melanoma cells into the dermal region appears to be associated with a localised disruption of the basement membrane. Presenting our results in terms of time course data, along with images and quantitative measurements of the depth of invasion extends previous 3D work that has often been reported without these details.