CHICAGO – Last month, healthcare and life sciences informatics startup Quantori introduced a high-performance computing accelerator meant to make HPC services for drug discovery and development more accessible for smaller pharmaceutical companies.
The Cambridge, Massachusetts-based firm simultaneously announced that the year-old Rectify Pharmaceuticals had adopted the technology for cryogenic electronic microscopy.
That is related to a paper that Quantori recently had accepted by Scientific Reports about its platform for drug discovery and development. That paper will describe how Quantori has helped researchers diagnose and identify the severity of COVID-19 by analyzing X-rays, according to CSO Yuriy Gankin.
Like other Quantori products, the HPC accelerator can be applied to genomics, ChIP-seq, RNA-seq, cell-image analysis, and computational chemistry.
Gankin likes to say that Quantori builds blueprints, accelerators, and platforms. Blueprints are essentially playbooks for deploying technology services, while accelerators are software modules that optimize performance on either cloud or high-performance computing infrastructure. Platforms have a more clinical application, such as diagnostics, genome annotation, and clinical trial management.
"Our goal, our strategy is to become a platform-based services company," Gankin said. "We don't want to become a product company, but we want to build modules and libraries and platforms which can be customized, which we can provide as a service."
Nor does Gankin want Quantori to become a transactional company. "We try to create partnerships. We don't do one-off projects unless they're pilots which may lead to long-term relationships," he said.
Quantori combines domain expertise in life sciences, software engineering, data engineering, and analytics, particularly in genomics and proteomics. Gankin noted that the company, which was founded shortly before the start of the pandemic, contributes to open-source communities in both cheminformatics and bioinformatics.
The company helps with infrastructure, laboratory informatics, early discovery development, and commercialization of drugs in Phase II trials and beyond.
Quantori's founders, namely Gankin, CEO Richard Golob, and Chief Technology Officer Vitaliy Aronov, have worked together for more than two decades. In 2001, the three started GGA Software Services, a bioinformatics outsourcing company they sold to Epam Systems in 2014.
The trio stayed at Epam, running that publicly traded firm's life sciences business development unit until they left near the end of 2019 to form Quantori in early 2020. Like GGA and the life sciences unit of Epam, Quantori provides bioinformatics and cheminformatics services to pharmaceutical and biotech companies as well as to major research institutions including the Jackson Laboratory.
Quantori has followed the strategy of first targeting early-stage but well-funded life sciences companies in the US, then slowly expanding to Western Europe market as well as to midmarket enterprises. In 2022, now that the firm has hundreds of employees, it is ready to take on pharma and biotech.
"We're finalizing delivery of major platforms and accelerators so that in 2023 we can position ourselves as a platform-based international company," Gankin said. Quantori may look at Japan, though it is not yet ready to make a full marketing push in Asia.
The company's customer roster now includes Pfizer, Takeda, PerkinElmer, Bruker, LGC, FogPharma, and Thrive Bioscience. About 80 percent of Quantori's business is with life sciences, with the rest coming from more general healthcare, Gankin said.
Quantori works with several venture capital funds, including Flagship Pioneering and Third Rock Ventures, to gain access to their portfolio companies as potential customers, though the company itself is still funded by its founders as well as by revenues.
"When we look for innovative startups, it doesn't make much sense to look for individual companies. Instead, we work with large venture funds and we ask them which companies may use our help," Gankin explained.
The firm employs about 850 people in North America and Europe, some of whom came over from Epam and GGA before that. Quantori has about 70 data scientists on staff, many with Ph.D.s, as well as several former pharma executives.
Gankin said that the total headcount will approach 1,000 by midsummer and that revenues would double in 2022, though the privately held firm does not release financials.
Quantori is in the business of integrative informatics, which brings together infrastructure, data structure, visualization, and the generation of insights. "Everything is well-structured data, which has to be processed, which has to be analyzed, and which has to provide insights," Gankin explained.
Quantori software runs on the three largest public clouds, namely Amazon Web Services, Microsoft Azure, and Google Cloud Platform. "Most importantly, [researchers] don't need help from professional software developers, so scientists can set up Jupyter [notebooks] with our systems themselves," Gankin said
Among projects Quantori has undertaken is the combination and normalization of genomic information, imaging data, and electronic medical records data with the help of accelerators and custom Python dialect, according to Gankin.
Another, which Quantori highlighted at the annual Bio-IT World conference in Boston last month, is its open-source development work with the Forome Association, an initiative that grew out of the Brigham Genomic Medicine program at Brigham and Women's Hospital. Forome, which stands for forum of omics, is an effort to crowdsource discovery of genotypic biomarkers that could be useful for diagnostics and drug development.
Quantori has helped the nonprofit develop a "lightweight," web-based data management and processing platform for omics analysis and literature mining, according to Michael Bouzinier, a senior research engineer at Harvard University who cofounded Forome.
The Forome Association started in 2018, when Bouzinier was director of informatics at Brigham and Women's, a position he held until April 2020. "I realized that the ambitions of the projects are not substantiated by the funding [the Brigham's] genomics lab had," he said. "We came out with the idea that maybe we should start an open-source initiative and find parties, organizations, and maybe individuals who are willing to contribute."
Forome came across Quantori while searching for partners, and the software firm is now the main contributor to the open-source community that has grown up around the technology, called Forome Platform. "Forome is our combined effort to provide a real platform for diagnosis and annotation of genomes," Gankin said.
Other members of the open-source community include Harvard, IBM, and Oracle.
"Our goal is to become the main diagnostics platform in several hospitals," with an initial focus on academic medical centers, said Bouzinier. The organization also wants to make a name for itself in the open-source community as the producer of a bioinformatics platform that is easy to install and manage.
The Forome Platform is designed for quick deployment in both local and cloud environments. Bouzinier said that support for on-premises installation is important because the organization is working with private data that a lot of research institutions do not want to expose to a cloud environment.
The technology developed for Forome is more rules-based than artificial intelligence or machine learning because the organization wants to focus on "explainability" of results, and AI is not that advanced yet, Bouzinier said.
However, he expressed hope that the next generation of Quantori technology would have "explainable" AI built in.
Gankin said that the model is less important than the utility of the technology. "Sometimes, some deep learning is required," he said.
One implementation of the Forome Platform, called Forome Annotation and Filtration for Sequencing Analysis (AnFiSA), produces visualizations to help clinicians without an informatics background understand Mendelian genetic variation. According to Quantori, Forome AnFiSA is meant to bring genetic analysis into routine clinical practice as well as to convince payors that genetic sequencing is medically necessary.
This technology also features cohort analysis for researchers looking for biomarkers for undiagnosed hereditary diseases.
Brigham Genomic Medicine has used Forome AnFiSA to diagnose patients who have lived with uncertainty for years. According to Quantori, the software helped the hospital detect Stickler syndrome in a woman in her 20s; the disease is typically diagnosed in early childhood.
Brigham researchers also have used AnFiSA to develop a protocol for detecting hearing loss in newborns as part of the Sequencing a Baby for an Optimal Outcome (SEQaBOO) research study.
However, Bouzinier said that it is too early for any Forome work to be translated to routine clinical practice, but that the organization has several scientific papers pending.