At the 6th H3D Symposium, PhD candidate Keaghan Brown presented a developing bioinformatics platform that aims to connect the genetic changes observed in viruses with their potential structural and functional effects — and, ultimately, inform early-stage drug discovery.
When a virus changes, the implications can extend far beyond the genetic sequence in which that change is first observed. A mutation may alter the structure of a viral protein, influence how it interacts with human cells or affect the way a potential drug binds to its target. Understanding these consequences is an important part of studying how pathogens evolve and how therapeutic strategies might respond.
Yet identifying a mutation is only one part of the picture. Translating genomic surveillance data into an understanding of what those changes might mean at the molecular level remains a considerable challenge in public health research and drug discovery.
It is this gap that PhD researcher Keaghan Brown is seeking to address through the development of a bioinformatics platform that brings genomic surveillance and structural bioinformatics together. He presented his research at the 6th H3D Symposium, held from 13 to 16 October 2026 at the Arabella Hotel, Golf & Spa in the Western Cape.
Rather than treating genomic data and structural biology as separate areas of investigation, Brown’s research explores how they can inform one another. The aim is to make information about viral mutations more useful for understanding pathogen behaviour, assessing potential implications for existing therapeutic interventions and prioritising candidates for further drug discovery research.
Looking beyond the sequence
Genomic surveillance has become an increasingly important part of monitoring pathogens. Through the sequencing of pathogen samples, researchers can identify genetic variation, track mutations and observe how viruses change over time. Platforms such as Pathoplexus provide valuable access to sequencing data and information about mutations, supporting the wider research community in monitoring pathogen evolution.
However, a genetic sequence alone does not necessarily reveal what a particular mutation means for a virus at a molecular level.
Some mutations may have little functional consequence, while others could affect the stability of a viral protein or alter the way it interacts with proteins in its host. Certain changes may also influence potential drug-binding sites, with implications for how existing or future therapeutic interventions could perform.
Brown’s research seeks to help bridge this gap. By bringing genomic information together with structural and molecular analyses, his work aims to investigate the possible consequences of viral mutations and connect those findings to the early stages of drug discovery.
The research focuses on Ebolavirus, a group of viruses that has been studied for decades but continues to present significant public health challenges. The opportunity, Brown explains, lies in making better use of the genomic information already being generated: moving beyond identifying that a virus has changed to investigating what those changes could mean.
Connecting four areas of analysis
At the centre of the research is a bioinformatics platform that brings together four areas of analysis: protein stability, virus–host interaction, allosteric binding and ligand screening.
Each component examines a different aspect of how mutations may influence viral proteins and their potential as drug targets. Together, they are intended to provide researchers with a more connected way of investigating the effects of genetic variation.
The process begins with mutation data obtained through genomic surveillance, which may come from clinical sequencing or resources such as Pathoplexus. Protein stability analysis then examines whether a mutation could affect the structure of a viral protein. This is important because changes to a protein’s structure can influence whether it continues to function as expected.
The platform also considers virus–host interactions: how mutations might affect the way viral proteins interact with proteins in the host. These interactions are important to understanding how viruses function within their hosts.
A third component, allosteric binding, explores potential drug-target sites away from a protein’s main binding site. These alternative sites can offer additional avenues for investigating how a protein might be targeted.
Finally, ligand screening examines potential compounds against selected targets. This helps researchers identify candidates that may warrant further investigation and laboratory testing.
The value of bringing these analyses together lies in the connections between them. Instead of examining a mutation in isolation, researchers can begin to trace its possible effects — from changes in a protein’s stability and interactions to the identification of potential drug targets and compounds for further study.
The platform is currently in development and integration, with a prototype nearing testing. Brown describes a workflow in which a researcher could begin with a mutation identified through surveillance, investigate its likely structural and functional effects, and use the resulting information to guide target selection and ligand screening.
Importantly, identifying a potential compound through computational screening does not establish that it will be effective as a treatment. Rather, the platform is intended to help generate hypotheses, prioritise promising candidates and guide the next stages of investigation.
What early findings could tell us
One preliminary insight from the research concerns viral proteins that appear to be particularly sensitive to mutation.
Brown explains that the team has identified proteins in which mutations frequently disrupt function. Because these proteins appear to tolerate relatively few changes, they may present a lower risk of drug resistance emerging through further mutation. This could make them particularly interesting candidates for targeting.
It is an encouraging observation, although it remains an early insight rather than a definitive conclusion. Further investigation and experimental validation will be important in establishing how these findings translate beyond computational predictions.
That distinction is central to the potential contribution of the platform. Computational approaches can help researchers narrow down where to focus their attention, but predictions need to be tested. Ligand screening can identify compounds worth pursuing; it cannot, on its own, establish that those compounds will work as treatments.
Incorporating experimental validation is therefore an important part of the platform’s development. The intention is to build a resource in which computational findings can be tested, refined and strengthened over time.
Why the research matters beyond the laboratory
The potential significance of this work extends beyond the technical development of a bioinformatics platform. It also speaks to the growing capacity for pathogen genomic research across Africa and the need to translate that capacity into further scientific opportunities.
African institutions have made considerable progress in expanding their sequencing capabilities. As more genomic data become available, tools that help researchers interpret the possible consequences of mutations could support more informed decisions about which viral proteins and pathways deserve further attention.
This is particularly relevant for diseases that affect African communities and may receive less investment in drug development than they require. By helping prioritise potential targets and compounds for laboratory testing, the platform could contribute to making early-stage drug discovery more focused and potentially less time-consuming and costly.
There may also be implications for public health research. By examining how mutations could affect viral proteins and their interactions, the platform could help researchers investigate variants that may have implications for transmission, pathogenicity or existing therapeutic strategies. Its role, however, is to support investigation and prioritisation, rather than independently determine whether a variant poses a public health threat.
Over time, Brown hopes this kind of work can contribute to building local capacity for computational drug discovery, enabling more African researchers to investigate pathogens affecting their own communities and participate in developing approaches to address them.
Taking the research beyond the symposium
At the 6th H3D Symposium, Brown presented the overall concept behind the platform, explaining how its different components connect genomic surveillance data to the likely structural and functional effects of mutations. He also demonstrated the prototype database, giving attendees an opportunity to explore how results could be viewed, and shared some of the team’s early insights.
The conversations that followed highlighted two priorities for the research community: accessibility and scientific validation.
Attendees were interested in whether the database would eventually be openly available, raising an important consideration for researchers who may have limited access to computational resources. There were also questions about how the platform’s computational predictions would be tested. Brown explained that the team plans to incorporate experimental validation to confirm key findings.
These discussions highlighted the expectations researchers have of resources intended to serve a wider scientific community: they should be accessible, and the findings they present should be supported by evidence. The feedback is also helping shape the platform’s priorities as the work moves towards broader testing.
Building a resource that lasts
Looking beyond the immediate development of the platform, Brown is thinking about how the database can remain useful beyond the duration of his PhD. A key priority is ensuring that the work does not lose momentum once he graduates.
In the near term, this means incorporating results from experimental validation and opening the database to contributions from other academics, allowing it to develop into a shared resource that can grow through wider research participation.
There are also plans to expand its coverage to other Ebolavirus species, particularly Bundibugyo virus, against the backdrop of the current outbreak. Extending the database in this way would allow researchers to investigate mutations in circulating strains and examine their possible effects on protein stability, host interactions and potential drug targets while an outbreak is unfolding.
Further ahead, Brown hopes the platform could provide a model that other research groups can adapt to pathogens beyond Ebolavirus, particularly those associated with neglected diseases in Africa.
The broader ambition is to demonstrate how genomic surveillance data can be connected to structural and functional insight, and how that information can then inform early-stage drug discovery. If the approach can be adapted to other pathogens, it could offer researchers a framework for making more effective use of the genomic data being generated in their own fields.
The work is still developing, with testing, validation and wider collaboration ahead. But its central idea is significant: understanding a pathogen requires more than knowing which genetic changes have occurred. It also requires investigating what those changes might mean — and how that knowledge can help guide the search for ways to respond.
