Identifying genetic variants that play a role in disease-related traits is challenging because of the complex relationships between the variants. Pinpointing these causal variants helps to develop new therapeutic targets or reveals new biological insights for diseases. Statistical fine-mapping aids this by identifying potential causal variants with the aim of reducing the number of genetic variants (i.e. improving the resolution) for follow-up in downstream functional validation experiments. As biologically related traits often have shared causal variants, fine-mapping traits together shares information between them and can give better resolution than fine-mapping each trait independently. There are few bioinformatics tools that summarise fine-mapping results, and those that do require some programming knowledge and are not interactive.

A new interactive visualisation tool, called flashfm-ivis (flexible and shared information fine-mapping – interactive visualisation), just published in Bioinformatics, provides a means of interactive exploration and publication-ready plots to summarise fine-mapping results; it is accessible without any programming knowledge.

Researchers from the MRC Biostatistics Unit (Feng Zhou, Jenn Asimit) and the British Heart Foundation Cardiovascular Epidemiology Unit (Adam Butterworth) developed flashfm-ivis to provide a simple way of exploring potential causal variants among traits.

Jenn Asimit says:

Plots are always the best way to get a good understanding of your results, and flashfm-ivis gives a point-and-click way to explore potential causal variants for disease-related traits. We hope that this tool will help others to more easily pinpoint variants for follow-up when trying to reveal the underlying mechanisms of disease.”

Flashfm-ivis allows users to view potential causal genetic variants that underlie associations that are shared or distinct between several traits and compares results between fine-mapping based on separate or joint analyses. It provides a series of plots that are all interactive with user-controlled zoom features.

One set of plots shows the evidence strength that each variant is causal for each trait – these plots are linked and allow users to select a set of variants in one plot and focus on these same variants in all other plots, allowing an easy comparison. Another plot shows the number of potential causal variants that are the same between different subsets of traits. There are two versions of these plots – one has segments that are proportional in size to the degree of overlap for an intuitive view and the other version allows users to view and download lists of variants in each combination of overlapping sets of variants between traits. Network diagrams show the potential causal variants that likely have joint effects on each trait and allow users to select a threshold for level of evidence to display, as well as the ability to move points around to change the perspective of the plot – see below for an example.

 

This research has been supported by the UK Medical Research Council, the Alan Turing Institute, British Heart Foundation, and the NIHR Cambridge BRC.

Feng Zhou, Adam S Butterworth, Jennifer L Asimit, flashfm-ivis: interactive visualisation for fine-mapping of multiple quantitative traits, Bioinformatics, 2022;, btac453, https://doi.org/10.1093/bioinformatics/btac453