DrugSense

DrugSense is a software tool able to predict the drug(s) to which the patient’s tumour will be most sensitive to by analysing the transcriptional profile of the tumour sample. The input of DrugSense is the transcriptional profile of the tumour, while the output is a list of drugs ranked according to their predicted potency in the patient, with the most potent at the top. DrugSense is based on a data-driven model derived from large publicly available collections of gene expression profiles of 1,000 cancer cell lines, and of dose-response curves of these cell lines to over 400 drugs and research compounds.

Publications

Computational drug prediction in hepatoblastoma by integrating pan-cancer transcriptomics with pharmacological response. Hepatology 2023. Abstract

Software guidelines

The following section provides guidelines on how to make better use of DrugSense.
DrugSense takes in input a CSV file that needs to be uploaded and an additional argument selected from a specific drop-down menu.
The input file contains normalized expression values for each gene in each tumor sample. The user must ensure that the file adheres to the CSV format, that genes and samples are placed on rows and columns, respectively, and that both gene and sample identifiers are provided (Figure 1). Note that the current version of DrugSense supports only Ensembl gene identifiers.

To get the best performance out of the algorithm, drug predictions for solid and liquid tumors are run separately. In this regard, it is recommended to pre-process data from solid and liquid tumors independently and to save the relative normalized expression profiles in separate files. In respect of data pre-processing, all the normalization methods suited for gene comparisons within a sample are well tolerated by the algorithm. In the following table are listed the most common one:

A specific drop-down menu (Figure 2) allows the user to select the tumor type corresponding to the expression profile(s) provided in input. This argument will instruct the algorithm on the drug-gene correlation profiles (DGCPs) to use; if no action is taken, the algorithm will import those trained on cancer cell lines from solid tumours, by default.

License Terms

Use of DrugSense tool is free to academic, government and non-profit users for non-commercial use. Use of DrugSense application for commercial use by non-profit and for-profit entities requires a license agreement. Parties interested in commercial use may first obtain a demo version of a stand-alone version of DrugSense application, subject to this License, by contacting Diego di Bernardo (dibernardo_at_tigem.it). Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

1. Redistributions of source code must retain the above copyright notice. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
2. The end-user documentation included with the redistribution, if any, must include the following acknowledgment: "This product includes software developed by the Systems, Synthetic and Computational Biology Laboratory of the TeleThon Institute of Genetics and Medicine (TIGEM)". If no such end-user documentation is to be included, this acknowledgment shall appear in the software itself, wherever such third-party acknowledgments normally appear
3. This license does not authorize the incorporation of this software into any proprietary programs.
4. THIS SOFTWARE IS PROVIDED "AS IS," AND ANY EXPRESSED OR IMPLIED WARRANTIES, (INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE) ARE DISCLAIMED. IN NO EVENT SHALL OUR INSTITUTE OR THEIR AFFILIATES BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.