NEW YORK – Nonprofit research organization Target ALS aims to speed the discovery of diagnostic biomarkers and treatments for amyotrophic lateral sclerosis (AML) with a free-to-access multiomic data portal.
The portal combines various data types, all derived from human samples, and makes them accessible to researchers, who can analyze everything on the cloud, via a suite of software services provided by both DNAstack and Verily.
Human samples for ALS, said Amy Easton, senior director of scientific programs at Target ALS, are "hard to come by, hard to get, and it's hard to make sure that whatever you get was well characterized and well preserved."
Target ALS and its partners, Easton said, had been growing their datasets over the course of several years, but lacked a streamlined way of allowing other researchers to access the data.
The portal currently contains data from the Target ALS Postmortem Tissue Collection, which consists of frozen and formalin-fixed, parrafin-embedded (FFPE) tissue from 300 ALS samples, which include several cases of frontotemporal dementia and 50 control cases. The collection of datasets includes semi-quantitative histopathology, whole-genome sequencing, bulk tissue RNA-seq, and spatial transcriptomics, where each spatial transcriptomics file includes spatial RNA-seq, histopathology images, and metadata.
The Target ALS data portal is powered by Omics AI, a bioinformatics software suite from Toronto-based DNAstack.
Marc Fiume, cofounder and CEO of DNAstack, said that Omics AI helps get around the problem of data siloization, while making the portal accessible and easy to use.
The status quo for data sharing, he said, has been to copy and move fairly sensitive information around the internet, while storing it locally.
"It's inefficient [and] it's a cost burden," Fiume said.
The Omics AI platform, he explained, provides researchers with a one-stop shop where data is stored and can be requested, queried, and, through a partnership with Verily, analyzed using the Verily Workbench, all on the cloud.
Fiume sad that DNAstack has collaborated with Verily over the past 10 years to develop a set of open standards aimed at solving the data siloization problem.
"We're really excited to continue to grow these datasets," he said, "and also to nurture the application ecosystem on top of [that]. By virtue of us building the platform in compliance with these standards, it allows scientists to create the next great analytical application, starting much further down the line than they would have otherwise been able to."
Users have only just begun to access the portal, so it remains too early to know specifically how researchers are using the data.
However, Yufeng Huang, a statistical geneticist at Biogen, said that from what he's seen of it so far, it looks like a promising tool for streamlining many tasks. Huang said in an email that his group has focused heavily on transcriptomic data analysis in the past, which requires internally storing large data files, but that if those can be processed "on the fly" with the Verily workflow set up, the entire process could be both faster and more cost effective.
"It would be quite exciting to use the data portal to access the whole-genome sequencing data for variant look-ups, sample retrieval, and potentially running genetic association analysis with some disease progression endpoints when the natural history data becomes available or future releases of the biofluid measures happen," he said.
Target ALS plans to update its portal every four to six months with data from an ongoing natural history study, as well as from stem cells collected from healthy individuals, patients with sporadic forms of ALS, and patients with known genetic mutations.
Target ALS launched a natural history study in 2021, with the goal of enrolling 800 ALS patients and 200 controls across the 18 participating sites worldwide. That study is set to run for five years and has so far recruited over 100 participants.
"The goal was to have this longitudinal biofluid collection [and] give it out freely for biomarker discovery," Easton said. "The second goal was also to to really help us understand what might cause ALS and in diverse patient populations."
Throughout the study, Target ALS is collecting longitudinal blood, urine, and cerebrospinal fluid samples, along with clinical data from participants every four months and up to 18 months. Data from this study and the stem cell collection effort will become available on the portal from May 1.
Easton said that the cause of ALS remains unknown in the vast majority of cases, but by longitudinally linking multiomic biological data to clinical assessments to understand patient histories and environmental impacts, a clearer picture of disease causes and progression should emerge.
"There probably is some genetic risk that we don't understand," she said, "and there's probably some environmental factor that we don't understand. We're going to get both of those pieces of information
Another important part of the natural history study, Easton said, is to gather data from patients of diverse ancestries, as the majority of research to date has taken place largely in populations of European ancestry.
Target ALS recently opened a site in Bogotá, Colombia and plans to open several more throughout South America, as well as in Israel and parts of Asia.
"A lot of the this work is important for industry," Easton said. "[Companies] can't conduct a proper clinical trial unless they have a biomarker that shows that their drug did what it was supposed to do."
Additionally, more and better biomarkers are needed for smaller, earlier clinical trials common throughout rare disease research, where it can be hard to detect a drug's benefit among a limited number of patients.
"We're super excited because we've got all of these unique data sets that don't exist in other places," Easton said. "Instead of making researchers ask for data [piecemeal], they can look at different data types from the same patient all in the same place."