Everything You Need to Know About the Open Targets Platform's Association Page

The Open Targets Platform (OTP) is an application that can be used to find and compare relevant data for the identification and prioritization of potential drug targets.  

One of the most important/informative pages in OTP is the association page. Here, a user can investigate multiple lines of evidence from a variety of data sources to find target-disease associations for a given target or disease. In order to contextualize the relative importance of each piece of evidence, the Open Targets Platform provides a scoring framework. This scoring framework aims to take into account differences in the data sources and reconcile these differences into a single score for every target-disease pair. This is done by taking the harmonic sum of all pieces of target-disease evidence belonging to that data source. The score is calculated using the next 3 steps (for details, see  the official OTP documentation):

• The evidence scores are sorted in descending order and assigned an incremental value that indicates their position in the sorted list (the top-scoring item has a positional id of 1, the second has a positional id of 2, and so on).
• The harmonic sum for each data source is then calculated by summing the result of dividing each evidence score by (positional id^2).
   

• To ensure the result is between 0 and 1, the harmonic sum is normalised by dividing the result by the maximum theoretical harmonic sum, which is the one calculated using an infinite vector of ones. The platform derives this calculation (which approximates to 1.644) by using a vector of 1,000 ones.

Finally, an overall association score for a target-disease pair is calculated by taking the harmonic sum of the association scores per data source weighted by the data source weights.

Now that we know how the association scores are calculated, let’s have a look at the association page for BRAF to see how a user can interact with it. The page contains a table with a disease on each row (indicating the target-disease pairs, where BRAF is the target). The columns contain the different data sources. The blue circles represent the association score of the data source for the specific target-disease pair. The blue squares in the first column represent the overall association score (Figure 1). Users can click on a blue circle to see the individual pieces of evidence that make up the association score. They can also change the weights of the data sources and apply some filters (Figure 2).

Additionally, each data source is implemented as an individual entity in OTP, making the association page very customizable. Developers with enough experience in working with OTP can add additional data sources based on public and private data.