Rocket Insights is helping accelerate Goldfinch Bio’s innovative approach to kidney disease treatment by automating the processing of large scale genomic and longitudinal patient data, streamlining complex workflows while reducing repetitive tasks, and creating efficient data visualizations for researchers to do better work, faster.
Data & Intelligence
Design & Technology
Product Design & Development
Goldfinch Bio is focused on discovering and developing precision therapies for patients with kidney disease. Chronic kidney disease (CKD) is a growing health epidemic affecting an estimated 37 million people in the U.S. (more than 1 in 7 adults). Over 23,400 Americans received a kidney transplant in 2019, while about 100,000 are still waiting for one. Dialysis and transplants are still very much the standard of care for most CKD patients, but there is a fast growing need for new treatment options to help this underserved patient population.
Over the past several decades, many companies have tried to develop new therapies to treat kidney disease, with few successes. But recent breakthroughs in genetics and biology have generated a greater understanding of the biological pathways that drive kidney disease.
Goldfinch Bio is translating these discoveries into new therapies that target the molecular causes of kidney diseases to help bring forth new treatment options for patients. The company has been collecting samples from affected patients, sequencing them, and enrolling them into a Kidney Genome Atlas™ (KGA) to try to build a data set large enough that the company can use statistical methods to identify potential drug targets with genetic evidence. The KGA represents the most comprehensive patient registry to investigate the underlying mechanisms of kidney disease.
Goldfinch was fortunate to launch a partnership with Gilead Sciences in 2019 to expand the scope of the KGA to include diabetic kidney disease. The strategic collaboration aimed to discover, develop and commercialize a pipeline of innovative therapeutics for diabetic kidney disease and certain orphan kidney diseases.
Goldfinch is a small company, with very limited engineering staff, but they had a strong desire to build a dynamic web-based application where they could surface all of the data they were collecting and producing as part of the KGA to help optimize target identification, patient stratification and the exploration of the relationship between genotype and phenotype.
Prior to this, Goldfinch’s translational medicine researchers and computational biologists were operating under a highly complex, disconnected workflow in order to be able to receive, process and analyze a massive amount of data coming from various sources, such as analyzed biopsies, tests, patient samples, etc. Goldfinch had multiple systems and software suites being used to look at the data, compare cohorts and test compounds in the lab. The company needed a way to create connective tissue between systems to be able to quickly go from curating to examining to validating their data.
“This is what really drove the project with Rocket - we wanted to make the data available to anyone in the company who could use it, and not have it locked up in a silo with the computational team who produced it.”
Rocket Insights was brought on to help augment the existing Goldfinch engineering staff and meet an aggressive timeline to launch an MVP of the web-based application. The project kicked off with a design sprint and moved quickly into development.
Following the design sprint, we did extensive user research amongst Goldfinch’s computational biologists and translational medicine researchers in order to better understand their workflow and what they needed from the data. From this user research, we clearly articulated the workflows that were needed for each team. We moved toward a vision of creating a software platform that could connect disparate workflows with visualizations that would make each team’s work quicker.
We quickly built a prototype of the platform so we could begin testing and iterating based on feedback from the research team. We learned that the interconnectedness of the platform was enabling the researchers and the biologists to take data, examine it and be confident in what they found, all in a much more efficient way. This was crucial, as it allowed the team to move faster in pursuing which gene mutations they wanted to pursue further research on.
From there, the prototype evolved into a powerful software application with a tremendous set of features and visualizations for the Goldfinch team.
Today, the KGA application built by Rocket is broken up into a few main sections:
These are analyses that are done using either public or private data. The app currently supports GWAS, differential gene expression, and rare variant analyses. Each type of study has its own interactive data viewer. Goldfinch is also hoping to add numerous other study types in the future, including single-cell RNA-seq, structural variants, epigenetic analyses, amongst others.
This is where the application surfaces longitudinal clinical data as well as medication data for all patients that have been enrolled in any KGA studies. This way users can track how certain kidney health metrics are tracking over time for any given patient and how that lines up with their medication schedule. It also includes lab data, social history, demographic information, family history of kidney disease, ICD codes, ESRD data, medication history and related comorbidities.
The Goldfinch team can explore the relationship between phenotype and genotype by applying clinical, demographic and genetic filters. These filters can be saved as a project and recalled for future exploration. Patients in a cohort can then be examined individually or from a high level such as demographic distributions and common genotypes. These views provide researchers information derived from trillions of patient alleles in seconds.
Genes and variants are surfaced pretty much everywhere in the app, but each has its own landing page where users can explore everything we know about it. This includes patient allele information, eQTLs, loss of function constraints, results from every study in the KGA, and other important metadata. We also provide convenience links out to many other public sites and databases.
This section is kind of like a shared collection of bookmarks into the human genome. We built a custom dashboard to help Goldfich’s scientific teams keep track of all of the genes and loci that are flagged as hits while analyzing public or private studies. These can also be added from literature or other sources and tracked in one centralized location. This makes it quick and easy to evaluate new potential targets from all of the data.
In the end, we worked with Goldfinch to bring together a wide array of data to enrich their core analyses and present as much supporting evidence as possible in one central location through the KGA app.
Goldfinch’s vision is to eradicate kidney disease through precision medicine. The ability to evaluate large amounts of data, automate processes with technology, and ultimately speed up their workflow is a big deal in reaching that vision.
Data is being generated at an exponential rate, and this won’t change anytime soon. But with the help of Rocket, Goldfinch is saving time by doing less manual work, and they are able to easily see complex data visualizations through the platform that would have previously required third-party tools and manual efforts to pull together. These visualizations allow the team to unearth important patterns that will inform further research and ultimately treatment in the future.
Going forward, Goldfinch is discussing new ways the company can do a better job of interacting with the patient community and getting them engaged with the KGA. Having a software platform that is end-user focused will be absolutely critical in achieving this, and we are glad to have played a role in creating that platform for Goldfinch.
“One of our employees had previous experience working with Rocket and we were very impressed by the breadth and depth of capabilities. They had no concerns about delivering on our timeline. Not only did we produce the MVP on time and on budget, but have continued the project, adding a tremendous set of features and capabilities to the application.”
- Adam Tebbe, Goldfinch Bio
Have questions about the project? We’re happy to answer any of your questions.CONTACT AN EXPERT