Skip to main content
Premium Trial:

Request an Annual Quote

Celmatix Relies on Informatics to Identify Biomarkers of Female Infertility for Diagnostic Test

Premium

By applying a series of algorithms to patients’ clinical and genomic data, New York City-based biotechnology firm Celmatix hopes to develop a noninvasive diagnostic test for female infertility that includes both phenotypic and genotypic biomarkers.

Specifically, the company's data scientists and biostatisticians are using machine-learning algorithms to build analytical models using clinical data from patients with unexplained infertility, including hormone levels and age, to understand how specific patients compare to other women with similar conditions.

The team is also analyzing whole-genome sequence data from its patient cohort in order to find predictive biomarkers that could be used in the diagnostic test, Piraye Yurttas Beim, the company’s founder and CEO, told BioInform in a recent interview.

The idea, she explained, is to eventually "integrate that genomic information into the predictive [phenotype] models that we are developing” and ultimately to create a test that helps physicians more easily interpret their clinical data, clarify why some patients are failing treatment, as well as guide them to more effective therapy strategies for their patients from the beginning.

The company is also studying premature menopause with the aim of eventually expanding the diagnostic to create a broader fertility screen that would be able to "identify at-risk women earlier in their lives before they become infertile and when they still have options such as egg freezing available to them,” Beim said.

As part of its research and development efforts, the company is partnering with fertility clinics to gather both genomic and de-identified clinical data from its target population. It is also working with sequencing services providers who are generating the genomic data it is using to develop the test.

Celmatix’s analytics team uses proprietary algorithms to streamline the identification of relevant patterns in genomic and phenotypic data, Beim said. To illustrate the process, she used the case of idiopathic infertility.

Women with this condition likely have several genetic factors at work, so “if we went to the doctors and said, 'Give us all your unexplained cases and we will sequence them all,' that’s a very noisy dataset,” she said.

However, the Celmatix team can use its analytics pipeline to identify which data from the patients should be fed into its biomarker discovery pipeline, she said.

On the phenotype side, “we use the analytics [tools] to model out the different treatment outcomes to determine which patients are truly idiopathic because even the best models applied to all phenotypic information physicians collect about these patients didn’t succeed in predicting that they would do as well or as poorly as they did, respectively, in treatment,” she explained. “That allows us to stratify our patient population in a way that reduces the noise in the genetic datasets that we are analyzing.”

The team is also seeking patterns in genomic data that could serve as biomarkers. As part of this effort, Celmatix’s bioinformatics team is developing a database called Fertilome, which it uses to identify relevant patient cohorts based on sequence data as well as to identify portions of the genome that include relevant biomarkers.

With the inclusion of a genetics component, “we are not just trying to find a deterministic SNP,” she said. Rather, “we want to [use] algorithms [to] … integrate multiple biomarkers,” enabling the company to provide information to clinicians that they would not get from analyzing phenotypic variables like hormone levels alone.

This is important because “if we only had a genetic biomarker discovery pipeline without the clinical data models, we might be excited by a very statistically significant biomarker but it would be hard for us to say …to the doctors: what extra level predictive power are you able to get from this genetic biomarker that your phenotypic variables don’t already give you?” she said.

By integrating genotypic and phenotypic data, "we can show that this is how far you can get with existing data [and] here is how much further you can go … if you actually incorporate genetic biomarkers,” she said.

Celmatix expects its diagnostic test to be ready for the market in the next two years, Beim said.

Filed under

The Scan

Genetic Tests Lead to Potential Prognostic Variants in Dutch Children With Dilated Cardiomyopathy

Researchers in Circulation: Genomic and Precision Medicine found that the presence of pathogenic or likely pathogenic variants was linked to increased risk of death and poorer outcomes in children with pediatric dilated cardiomyopathy.

Fragile X Syndrome Mutations Found With Comprehensive Testing Method

Researchers in Clinical Chemistry found fragile X syndrome expansions and other FMR1 mutations with ties to the intellectual disability condition using a long-range PCR and long-read sequencing approach.

Team Presents Strategy for Speedy Species Detection in Metagenomic Sequence Data

A computational approach presented in PLOS Computational Biology produced fewer false-positive species identifications in simulated and authentic metagenomic sequences.

Genetic Risk Factors for Hypertension Can Help Identify Those at Risk for Cardiovascular Disease

Genetically predicted high blood pressure risk is also associated with increased cardiovascular disease risk, a new JAMA Cardiology study says.