Illuminating a dark target
Finding an appropriate dark target to study
Exploring potential compounds to affect a rare disease
Exploring potential effects of a novel chemical compound
Exploring commonalities amongst a set of ligands identified by a screen
Use Cases
Pharos has many features to aid in knowledge discovery and hypothesis generation. The use cases highlight how Pharos' individual features can be used together to reveal patterns in the data.
Illuminating a dark target

This use case profiles the features of Pharos that help a user begin to understand a dark target, and generate hypotheses for its role. After reviewing the primary documentation for that target, the dataset is expanded to a list of interacting targets. The tutorial shows you how to do enrichment analysis on the list, and create a heatmap of data for the list. The goal is to highlight patterns in the properties of a set of related targets to help build hypotheses about the role of the dark target.

A biologist is studying rare diseases. Based on some results of a recent GWAS study, she would like to investigate potential roles of a target in a rare disease, and potential medical interventions to affect the course of the disease.

She begins by finding her dark target, and reviewing primary documentation for it.

Example Dark Target

As you might expect, there is not a lot of primary documentation for her target. She finds no other associations to the disease, no significant GO Terms, and no documented involvement in relevant Pathways. She did find several protein-protein interactions pulled from the STRING-DB database, however. Perhaps the interacting proteins have relevant documentation.

Example Target List

The resulting target list close to 150 targets, which are documented to be associated with several diseases, pathways, and GO Terms. She notices the list of associated diseases includes many types of cancers, which was not expected. She knows that cancer is very well studied, and wonders if the number of targets in the list that are associated with cancer is greater than would be expected by random chance.

Example Enrichment

The filter value enrichment shows that cancer is not actually over-represented in the list, meaning the high number for targets associated with cancer may just be an artifact. Calculating enrichment for some other filters yields a few interesting GO Terms and Pathways that could look into. She also creates a heatmap of target-ligand activity for the interacting targets.

Example Heatmap

There are a few compounds that could be used to perturb the system, and potentially affect the course of the disease. She downloads the target-ligand activity data for further investigation and to see which compounds she might be able to get ahold of.

Finding an appropriate dark target to study

This use case highlights the features of Pharos that would help a researcher find a dark target to study based on another target they know well. The tool shows how to find a set of similar targets based on protein sequence, or common documentation. The resulting set can be filtered based on the Target Development Level to highlight dark proteins. Other relevant filters can help highlight targets that have available IDG mouse models or genetic constructs, or have an ortholog in the researcher's preferred model system.

A grad student in a Neuroscience Department wants to design a research project for his thesis.

The lab he works in has a lot of experience and equipment devoted to the study of calcium channels in mouse models. A recent publication from the lab added to what is known about CACNG1, a regulatory subunit of a calcium channel. He begins, as usual, by reviewing primary documentation for CACNG1.

Example Dark Target

In Pharos, the student generates a set of similar targets based on a sequence search for targets related to CACNG1. He also could have generated a list of targets from the same DTO class, or PANTHER class, to find related proteins.

Example BlastP List
Example Filtered List

He finds 7-8 related targets. After highlighting the 'Tdark' proteins by filtering the list, he notices some have IDG resources, specifically 3 cell and 1 mouse resource he could potentially use for his project.

Example Enrichment

He also looks into the associated diseases of his dark target, and similar targets, and proposes some future studies into the role of his target in disease. See Use Case 'Illuminating a dark target" for more.

Exploring potential compounds to affect a rare disease

This use case profiles the features within Pharos to investigate knowledge about a disease. Researchers can explore patterns in the targets that are known to be associated with the disease. Furthermore, generating heatmaps of those targets versus all the compounds with activity against those targets can reveal patterns in the dataset. Researchers can also download data for all the associated targets and their active compounds for further analysis.

A biologist studying a rare disease wants to review compounds that could potentially affect the progression of the disease. Starting at the disease details page, they explore the full list of associated targets for that disease.

Example Disease

The list of associated targets includes many proteins and they want to narrow the search to targets that have been found to be upregulated in the disease.

Example Filtered List

They browse all Pharos' ligand activity data for the targets in the list in a heatmap. They sort by different columns to find highly potent compounds for some of the targets. They download the ligand activity data for more follow up.

Example Heatmap
Exploring potential effects of a novel chemical compound

This use case highlights the tools within Pharos to study the potential effects of a novel chemical compound. Pharos has the ability to generate a Target List of targets that the compound is predicted to have activity for. Alternatively, generating a Ligand List of compounds with a similar structure can help the researcher understand which compounds may be selective, and which targets, target classes, or target pathways can be affected by the compounds in the list. Highlighting these patterns in a list of similar compounds can help understand the potential effects of the novel compound.

A research chemist has a novel chemical compound that could potentially be useful as a therapeutic agent. To begin investigating potential effects and off-target effects, she performs a series of structure searches on Pharos.

First, she searches for predicted targets for the structure.

Example Predicted Targets

After finding a few interesting targets to follow up on, she also decides to characterize potential effects by studying the activity profile of similar compounds. The chemist notices that the list of similar compounds has some interesting patterns of active targets, and target classes. Additionally, the target count histogram tells her which ligands are known to be promiscuous in the targets they affect, and which may be selective.

Example Similar Compounds
Exploring commonalities amongst a set of ligands identified by a screen

This use case highlights the tools within Pharos to analyze a list of chemical compounds identified by a screen. Loading a list of compounds into Pharos can be done with a number of identifiers including SMILES, ChEMBL IDs, etc. The features and visualizations available on the resulting Ligand List can help the researcher understand which compounds may be selective, and which targets, target classes, or target pathways can be affected by the compounds in the list.

A chemist has screened a hundred thousand compounds from their library against a cell culture screening assay designed to identify compounds that affect a particular cellular process. The screen has identified about 500 compounds that have a reliable affect on the measured behavior.

He uploads the list of compounds into Pharos for analysis, using the SMILES for the identified compounds.

Example Custom Ligand List

Analyzing the list provided some useful insights. The chemist was able to identify some common active targets for compounds in the list. The list was also enriched in compounds with activity towards targets in a specific Reactome Pathway. It also proved useful to identify some relatively selective compounds using the Target Count filter.

Example Enrichment