The detailed illustrated user guide in PDF format (including a case study) can be downloaded here.

Workflow for DrugTargetInspector:

DTI workflow overview
Clicking on a box in the workflow image opens the respective section in the tutorial below.

Step-by-step tutorial for DrugTargetInspector:

DrugTargetInspector offers you three variants for data input. You either may choose to import preprocessed and normalized expression data from GEO, provide a precomputed list of scores or upload a gene expression matrix.

Gene Expression Omnibus

The Gene Expression Omnibus (GEO) is a MIAME compliant online database for microarray experiments. Normalized data is stored in the GEO SOFT format, whereas unprocessed data is stored in a platform dependent raw format.

When using a record from GEO, DrugTargetInspector relies on the proper normalization of the stored data. If you want to normalize the data yourself, you will need to obtain and process the raw data from GEO and upload a score file.

The SOFT format is supported for GEO Datasets (GDS) and GEO Series (GSE). DrugTargetInspector requires you to select either one GSE record and distribute the contained samples into a test set and control set or select two GDS records that define your sample and reference set.

In case you choose a GSE file enter a valid GSE identifier (e.g., GSE14767). The corresponding GEO Series .soft file is then downloaded to the DrugTargetInspector server automatically. In a next step, you may specify the sample and the reference group.

In case you choose two GDS files enter valid GDS identifiers (e.g., GDS2161 and GDS2162) for the test and control group, respectively. The cor­re­spon­ding GEO Data Set .soft files are then downloaded to the DrugTargetInspector server automatically.

In case you choose a text file upload a plain text file containing identifier with pre-computed scores or representing a gene expression matrix. The values have to be whitespace- or tab-delimited. (example)
If you have uploaded a GSE file or a gene expression matrix containing both, data of the test group and the control group, the sample identifiers (GSMs) / samples are displayed in the data pool.

You can then select arbitrary GSMs / samples and move them either to the sample group or to the reference group. DrugTargetInspector also provides a link to inspect the GSE file on the NCBI webserver.

In this step a score for differential expression between the two groups is calculated.

If your test group consists only of a single sample (e.g. for diagnostic purposes), DTI offers the following schoring methods:

  • Z-Score
  • Log-Mean-Fold-Quotient
  • Mean-Fold-Quotient

If however your test group consists of multiple samples you can choose from the following scoring schemes:

  • Independent Shrinkage t-Test
  • Independent Student's t-Test
  • Wilcoxon Rank Sum Test
  • Signal to Noise Ratio
  • F-Test
  • Log-Mean-Fold-Quotient
  • Mean-Fold-Quotient

Details on the mathematic formulations of the single tests can be found here.

Select a .vcf file from your local file system for upload and annotation. Click Browse to select a file and Upload mutation data to upload and annoate the file using Ensembl's Variant Effect Predictor (VeP).

The annotation process might take several minutes.

You can also continue to DrugTargetInspector without uploading a file by clicking Not now.

For testing purposes, you can use an exemplary VCF file by clicking Use example .vcf file.

You can select the subtype of the tumor under investigation from the dropdown menu. Potential targeted treatment options recommended for a given subtype are then listed on DrugTargetInspector's result page.

The list of cancer subtypes and corresponding treatment options are obtained from the American Cancer Society (ACS).

You can also continue to DrugTargetInspector without selecting a tumor type by clicking Not now.

The sortable table contains information on all significantly deregulated genes, which also are drug targets according to DrugBank.

The significance threshold is computed as mean +/- standard deviation.

links to PubMed search results for the respective drug and corresponding gene.

The Target Name column can be sorted alphabetically and the Score column numerically by clicking on the corresponding header. The arrows indicate the current sorting scheme, in this case according to scores in decreasing order.

The number of table rows per page can be set using the dropdown menu in the top left corner.

The table can be searched for target and drug names, as well as scores using the interactive search field on the right side.

Clicking the Print view button creates an easy-to-print version of the currently visible table content.

Clicking on the target name opens a new tab with the corresponding molecule's entry in NCBI Gene.

Clicking on the name of a drug links to its database entry in DrugBank. links to PubMed search results for the respective drug and corresponding gene.

The side panels contain settings and analyses applicable to the complete results table.

The Filter side panel allows to filter drugs according to their degree of deregulation (Show only deregulated drug targets), as well as pharmacological properties like being an antineoplastic agent and / or an inhibitor.

The content of the remaining side panels is detailded in the following sections.

Once a tumor subtype was selected (either in the wizard or in the Treatment recommendations side panel), the side panel displays links to treatment information on the websites of the American Cancer Society.

In cases of subtypes for which targeted treatment options are available, the respective drugs are listed in the side panel, as well as highlighted in the results table in lime green.

Using the checkbox Show only recommended drugs, the user can toggle between the complete list of all drug targets and a table containing only those molecules that can be targeted by recommended drugs.

In cases where the activation of the Show only mode results in an empty table, the user should consider the settings in the Filter side panel and for example activate the display of all drug targets, not just the deregulated ones.

The computation of enriched KEGG pathways containing a drug target of interest can be started by first clicking on the and then on the Show KEGG enrichment button.

The results are listed in an additional side panel on the right.

The side panel lists all KEGG pathways containing the drug target under investigation. Colored arrows indicate whether a certain pathway is significantly enriched (red arrow pointing upwards) or depleted (green arrow pointing downwards).

Clicking on the pathway name links to the pathway description on the KEGG website. Clicking on the K button opens the KEGG visualization of the pathway with the considered drug target highlighted in red.

The button View enrichment results forwards the user to a detailed list of investigated pathways, elucidating furhter properties of the results like the p-value, the genes contained in the pathway and running sum plots.

Besides a KEGG-specific gene set enrichment analysis, the user can perform general gene set enrichment analyes. Clicking on the Select enrichment algorithm button in the Gene set enrichment side panel opens a dropdown menu, where the user can choose between several enrichment algorithms.

The user can then select the biological categories the enrichment is computed on from a wide range of predefined categories or upload own categories in .gmt format.

In a next step, parameters for the computation can be set. However, the default settings are carefully chosen and do not require further adjustment in most cases.

After the computation is finished, the user is forwarded to the results page. There the results from all computations performed during the current session are listed. The first entry is the result of the most recent computation, in this case the gene set enrichment.

Clicking on the View button results in a set of collapsable panels that represent the overall categories as selected before.

Clicking on a collapsable lists the details for the respective categories: whether the category is signifcantly enriched (red) or depleted (green), and its p-value. The link More... opens an additional window with even more fine-grained information like the list of contained genes or the running sum that was created during the computation.

In order to compute the most deregulated subgraph with a drug target of interest as root node, a subgraph analysis has to be performed [1] . Clicking on the and then on the Compute subnetwork button starts the computation based on the settings made in the Subgraph analysis side panel (see next tab). After computation, a .jnlp file will be downloaded, which - when clicked on - opens BiNA [2] in its Webstart version to display the resulting subgraph.

In cases where the considered molecule is not contained in our underlying KEGG network or no subgraph of predefined size is available, the Compute subnetwork button is shown grayed and inactive.

In the Subgraph analysis side panel, the range for the computation of subgraphs can be determined. In order to assess the robustness of the results, we recommend to compute subgraphs of several sizes (as described by the range). The resulting subgraphs are then displayed in a consensus view in BiNA.

Besides the range, the user can select whether to compute a graph that lies downstream of the considered drug target or upstream of it.

Also, one can choose between the most up-regulated subgraph (the absolute values of the computed scores ares considered) or the most de-regulated subgraph (scores stay untouched).

The subgraph computation is started from the table as described in the previous tab. In cases where the overall most deregulated / upregulated subgraph (which is not rooted in a predefined node) is of interest, the button Start NetworkTrail analysis links to a new page where the user is guided through the settings for this general type of analysis.

After computation of the subgraphs, a .jnlp file is downloaded. In order to view the results, the file has to be opened and possible security warnings have to be acknowledged.

The visualization tool BiNA is then opened as a Webstart, which can take a few moments at the first startup.

The results of the subgraph computation are then displayed in BiNA. The scores of deregulation are mapped on the corresponding genes / proteins in the network, red indicating up-regulation and green down-regulation. Pill symbols indicate that the flagged protein is a drug target.

Do you have problems starting BiNA? Please refer to our troubleshooting help page.

Once mutation data was uploaded to DTI (either in the wizard or via the Mutation data side panel), the results table contains an additional column indicating whether (blue icon) or not (gray icon) a gene is affected by a mutation. In cases of existing mutations, a click on the blue map icon opens a popup which contains details on the effects of the mutations. The transcript IDs are linked to their corresponding entry on the Ensembl homepage.

The Mutation data side panel contains a legend on the genomic positions of various types of genomic effects, as well as statistics on the uploaded mutation dataset.

All subgraph anlaysis results will be extended by the visualization of existent mutations in the genes contained in the subgraph. The affected genes are highlighted with a blue border.

All subgraph anlaysis results will be extended by the visualization of existent mutations in the genes contained in the subgraph. The affected genes are highlighted with a blue border. Clicking on such a gene zooms-in to its position in BiNA's integrated genome browser. Zooming and searching within the genome browser allows to cover several layers of granularity.


  1. Backes, Christina and Rurainski, Alexander and Klau, Gunnar W and Müller, Oliver and Stöckel, Daniel and Gerasch, Andreas and Küntzer, Jan and Maisel, Daniela and Ludwig, Nicole and Hein, Matthias and others An integer linear programming approach for finding deregulated subgraphs in regulatory networks Nucleic acids research Oxford Univ Press
  2. Gerasch, Andreas and Faber, Daniel and Küntzer, Jan and Niermann, Peter and Kohlbacher, Oliver and Lenhof, Hans-Peter and Kaufmann, Michael BiNA: a visual analytics tool for biological network data PloS one Public Library of Science