Skip to article frontmatterSkip to article content

Spotlight

Sun’s series work in drug discovery

Jianfeng Sun spearheaded a research plan dedicated to the AI-based discovery of small molecule therapeutics targeting non-coding RNAs and proteins, working in collaboration with both experimentalists and computational scientists across the globe. He has released 5 studies as in Table 1. The development of DeepdlncUD was one of the series studies of this project.

Table 1:Sun’s work in drug discovery. ➵ stands for the current work.

FieldMoleculeTool nameFunctionTechnologyPublication
Systems Biologynoncoding RNADeepsmirUDdrug discoveryArtificial intelligenceSun et al., 2023. International Journal of Molecular Sciences
➵DeepdlncUDdrug discoveryArtificial intelligenceSun et al., 2023. Computers in Biology and Medicine
proteinDrutaidrug discoveryArtificial intelligenceSun et al., 2023. European Journal of Medicinal Chemistry
Structural BiologyproteinDeepHeliconstructural predictionArtificial intelligenceSun and Frishman, 2020. Journal of Structural Biology
DeepTMInterprotein-protein interaction predictionArtificial intelligenceSun and Frishman, 2021. Computational and Structural Biotechnology Journal

Feature of the computational method

Our approach centers on a computation-led framework that reconstructs small molecule–ncRNA relationships as an alternative to traditional count matrix-based differential expression (DE) analyses. Instead of relying on DE results, we predict the upregulation or downregulation of ncRNA expression mediated by small molecules. For each small molecule, genes are first categorized into upregulated or downregulated groups, and then ranked within each group based on the predicted likelihood of expression change. Building on our initial instance in applying this framework to identify candidate drugs targeting miRNAs , we extended its applicability by integrating DeepdlncUD, a tool specifically designed for lncRNA targets. The drug-like molecules inferred through this approach hold promise for both drug repositioning and the discovery of novel therapeutics. To the best of our knowledge, this represents the first fully automated, computation-driven drug inference pipeline based on connectivity scores that relies solely on molecular sequences to accomplish an end-to-end discovery process.

Support for running multiple cases

In this updated version 0.0.1 in PyPI, we tuned DeepdlncUD to make it available to run multiple instances of predicting SM-lncRNA regulation types. This should be seen as a major update because it is very important for researchers to screen large-scale regulation types with reduced oprations from their back ends.

Runtime

DeepdlncUD was developed based on molecular sequences alone. It runs in a very fast speed than those supported with complex data structures and settings, making it an ideal tool for biochemical researchers.

References
  1. Sun, J., Ru, J., Ramos-Mucci, L., Qi, F., Chen, Z., Chen, S., Cribbs, A. P., Deng, L., & Wang, X. (2023). DeepsmirUD: Prediction of Regulatory Effects on microRNA Expression Mediated by Small Molecules Using Deep Learning. International Journal of Molecular Sciences, 24(3). 10.3390/ijms24031878
  2. Sun, J., Si, S., Ru, J., & Wang, X. (2023). DeepdlncUD: Predicting regulation types of small molecule inhibitors on modulating lncRNA expression by deep learning. Computers in Biology and Medicine, 163, 107226. https://doi.org/10.1016/j.compbiomed.2023.107226
  3. Sun, J., Xu, M., Ru, J., James-Bott, A., Xiong, D., Wang, X., & Cribbs, A. P. (2023). Small molecule-mediated targeting of microRNAs for drug discovery: Experiments, computational techniques, and disease implications. European Journal of Medicinal Chemistry, 115500. https://doi.org/10.1016/j.ejmech.2023.115500
  4. Sun, J., & Frishman, D. (2020). DeepHelicon: Accurate prediction of inter-helical residue contacts in transmembrane proteins by residual neural networks. Journal of Structural Biology, 212(1), 107574. 10.1016/j.jsb.2020.107574
  5. Sun, J., & Frishman, D. (2021). Improved sequence-based prediction of interaction sites in α-helical transmembrane proteins by deep learning. Computational and Structural Biotechnology Journal, 19, 1512–1530. 10.1016/j.csbj.2021.03.005
  6. Samart, K., Tuyishime, P., Krishnan, A., & Ravi, J. (2021). Reconciling multiple connectivity scores for drug repurposing. Briefings in Bioinformatics, 22(6). 10.1093/bib/bbab161
  7. Arun, G., Diermeier, S. D., & Spector, D. L. (2018). Therapeutic Targeting of Long Non-Coding RNAs in Cancer. Trends in Molecular Medicine, 24(3), 257–277. 10.1016/j.molmed.2018.01.001
deepdlncud
Publication
deepdlncud
Versions