Genomenon Scores $1.7M SBIR Grant to Curate Genetic Drivers of Diseaseled post
Genomenon announced Tuesday that it was awarded a $1.7 million Small Business Innovation Research (SBIR) grant from the National Institutes of Health to help it organize and connect patient DNA to global genomic research in the areas of rare genetic diseases and cancer.
The grant is the latest in a steady stream of company advances highlighted by the addition of a chief financial officer in January and several collaborative agreements in the past year.
“The SBIR grant accelerates the development of our artificial intelligence (AI) engine to curate the genetic drivers for each disease,” Mike Klein, CEO of Genomenon, told BioSpace. “This enables Genomenon to put actionable genomic data at the fingertips of clinicians who are diagnosing and treating patients with genetic diseases and cancer, and into the hands of researchers developing precision medicine.”
AI, increasingly, is becoming a mainstream, must-have tool for drug developers, letting them analyze massive quantities of data quickly and accurately and find and make correlations that lead to meaningful insights far sooner than otherwise would be possible if such insights were made at all given the vast quantity of data.
Yet, Klein said, “There is no out-of-the-box, one-size-fits-all solution for AI, especially in genomics. For example, Natural Language Processing (NLP) simply doesn’t work in finding and extracting genomic data because there is nothing natural about the way genomics is described in the scientific literature.” Consequently, data interpretation often may be incomplete, inefficient and laden with errors because it relies upon largely manual activities.
“Developing AI for genomics is a pain-staking process that required us to start with the fundamentals of molecular biology and build a bespoke set of algorithms specifically around the science,” he said. Genomenon found a way to automate the search and to combine expert curation and machine learning methods that can always incorporate the latest, most relevant information.
Genomenon developed the Mastermind® Genomic Search Engine to improve and accelerate diagnostics within genetic testing labs. Then it developed the Mastermind® Genomic Landscapes application to deliver precision medicine developments and genomic biomarkers for clinical trial target selection and, therefore, to provide the hard evidence to support companion diagnostics development. Its goal is to curate the entire human genome.
The current iteration of the Mastermind Genomic Search Engine contains all the disease-gene-variant-phenotype-therapy relationships found across the scientific literature, which would make it the most complete source of genomic information available to medical researchers. It is used by some 1,000 diagnostic labs and is integrated into 18 clinical-grade support platforms and reference databases worldwide, enabling physicians to connect their patients’ genetic variants to published genomic researchers.
This linkage is particularly beneficial for patients with cancer and those with rare diseases. For example, in rare diseases, information is often lacking during the early stages of diagnosis. Nearly 80% of rare diseases have a genetic element, but those elements tend to be poorly characterized.
This lack of information slows diagnosis, stretching it out over the years, and affects treatment decisions. Knowing the implications of specific variants in a patient’s genome helps therapy begin sooner and gives clues as to which treatment may be most effective. For drug designers, the insights enabled by comprehensive genomic searches will help them optimize drug design, development and delivery.
To those ends, this SBIR grant will extend the functionality of the Mastermind Genomic Search Engine by adding expertly curated evidence and collaborative community features. As a result, clinicians can better interpret variants and therefore make more informed genetic diagnoses.
“Having already curated more than 250 genes for rare diseases and cancer, this NIH funding allows us to accelerate the pace of genetic interpretation for our AI-driven curation engine, which is a crucial first step in a process that enables targeted and efficient use of expert human curators,” Klein said in a statement. “Next-generation sequencing gives us the
ability to quickly and efficiently create a large amount of genomic data. Curating the entire genome for meaning and actionability at scale is the next step in the evolution of using this information to improve patient care.”
Significantly, Genomenon’s search engine algorithms optimize the review process by prioritizing search results based upon relevance. This process is called the Mastermind Relevancy Score. The search engine reviews the latest research first to return a score, ascertaining clinical relevance and then contextual relevance. Clinical relevance considers the type of journal, its journal impact score, genomic-specific information, and publication date. Contextual relevance relates to the specific search being launched. It is based upon the likelihood of the article focusing upon – rather than merely mentioning – the gene or variant of interest and the strength of association between a disease and a specific genetic element.
Searchers can affect prioritization by adding specific disease terms, phenotypes or a variety of search subcategories to the search.
Genomenon partnered in January to integrate Mastermind® into OmniTier’s CompStor® genomics analysis platform. A few months earlier, it announced a partnership with Inozyme Pharma to diagnose and make treatment decisions regarding patients with the rare ENPP1 deficiency. It also is working with Deep 6 AI, Alexion, AstraZeneca Rare Disease and Nostos Genomics.
“Our next inflection point is a fully curated clinical exome,” Klein said. “We’re getting there gene-by-gene and disease-by-disease, and expect to have this in hand in the next two years.”
View source version on biospace.com: https://www.biospace.com/article/genomenon-receives-1-7-million-sbir-grant-to-extend-its-ai-genomics-engine-/?s=72