https://github.com/SVAI/Gleukos/blob/master/nf.ipynb
We use data science approach by leveraging the drug screening data to reveal enriched pathways in treating plexiform neurofibromas. Our resuts suggested novel pathways for this disease and some drug combinations for the treatment.
Plexiform neurofibromas (pNFs) is a serious type of Neurofibromatosis. It impact ~40% of people with Neurofibromatosis Type 1 (NF1). Other than the other types of Neurofibromatosis, this subtype can lead to death. In this work, we use a data science approach to suggest some drug combinations for this disease.
Based on the drug screening data provided, we first linked it to a more general CHEMBL dataset of drug targets. Then many relevant targets are revealed for the compounds which showed at 1 micromolar potency to the pNFs cell lines. We further analyzed the associated pathways of the relavent targets and obtained 10 enriched ones. From the enriched pathways, we can select drug combinations that are more robust in treating pNFs.
We observed 10 enriched pathways as shown below. Some are very interesting because in addition to some cancer-related pathways, we found pathways like proteosome, calcium signaling, thyroid hormone. Those provide us with novel targets to consider for drug discovery.
(https://github.com/SVAI/Gleukos/blob/master/output_18_1.png)
By integrating the drug screening data and multiple datasets, we found several pathways associated to the plexiform neurofibromas. We also provided some drug combinations suggestion for the biological assays next step.
None. All are uploaded on this repo.
The next steps would do drug safety check and do animal model experiments.
Drug development pipelines like assays and animal model.
Experimental skills to test our prediction.
You can reproduce all the findings by running https://github.com/SVAI/Gleukos/blob/master/nf.ipynb