Today's Clinical Lab - News, Editorial and Products for the Clinical Laboratory
A 3D illustration of chromosome and cell nucleus with telomere and DNA representing gene therapy or cytogenetics.
AI may potentially improve data analysis in cytogenetic research, which examines changes in chromosomes and is primarily used in prenatal, postnatal, and cancer research.
iStock, wildpixel

Accelerating Cytogenetics Research with AI

Combined with automated data analysis, AI could increase clinical research labs’ efficiency and productivity

Photo portrait of John Burrill
John Burrill, PhD

John Burrill, PhD, is director of product management and market development, reproductive health, at Thermo Fisher Scientific. Burrill joined the company as director of portfolio and program management in March 2016 through Thermo’s acquisition of Affymetrix. At Affymetrix, he served in a variety of roles, the last being as senior director for software development, leading the team responsible for Affymetrix commercial software. Prior to joining Affymetrix, Burrill worked at Applied BioSystems and Incyte Genomics. Burrill is a graduate of the University of Chicago and holds a PhD in neuroscience from the University of Michigan.

ViewFull Profile
Learn about ourEditorial Policies.
Published:Jul 05, 2023
|3 min read
Register for free to listen to this article
Listen with Speechify
0:00
5:00
Photo portrait of John Burrill
John Burrill, PhD, is director of product management and market development, reproductive health, at Thermo Fisher Scientific. Burrill joined the company as director of portfolio and program management through Thermo’s acquisition of Affymetrix. At Affymetrix, he served in a variety of roles, the last being as senior director for software development, leading the team responsible for Affymetrix commercial software. Burrill is a graduate of the University of Chicago and holds a PhD in neuroscience from the University of Michigan.

In recent years, laboratories have undergone transformation through advances in automation and digitalization driven by the rise of artificial intelligence (AI). One could say that the lab of the future is here.

While AI promises to accelerate and improve the interpretation and reporting of genetic research, there’s also room for innovation at the front end of the workflow. Advances in genetic analysis technology will improve data quality and turnaround times, and facilitate more scalable, cost-effective research. Combined with automated data analysis, these technological innovations could help clinical research labs greatly increase efficiency and productivity.

Automating the interpretation and reporting of genetic research

AI holds a distinct promise for improving data analysis in cytogenetic research, which examines changes in chromosomes and is primarily used in prenatal, postnatal, and cancer research. Chromosome banding analysis is the gold standard to identify cytogenetic abnormalities, and AI can greatly increase the efficiency of what was once a very time-consuming and complex task.

With the addition of AI to digital automation, labs can now quickly analyze data from genetic research studies, accurately interpret genetic information, and turn it into meaningful results. Manual interpretation required hours of research to find relevant pathogenic associations. Now, with automated data interpretation and reporting, labs can almost instantaneously identify the most important mutations in a sample, greatly increasing productivity.

Exploring emerging technology applications

As the cost and complexity of next-generation sequencing (NGS) continue to decline, researchers are increasingly exploring the use of NGS-based approaches, including exome sequencing (ES) and whole genome sequencing (WGS) for cytogenetic analysis. However, while NGS has begun to revolutionize other research applications, the technology is still evolving for cytogenetic analysis and comes with challenges:

How Are Genetic Variants Classified?
Genetic variants can be benign or likely benign, or classified as pathogenic, likely pathogenic, or labeled inconclusive, i.e., a variant of uncertain significance (VUS). Clinical genetic testing reports on pathogenic, likely pathogenic, and VUS variants according to ACMG/AMP guidelines.

WGS provides massive amounts of data, which can complicate interpretation, given the increased likelihood of coming across variants of unknown significance (VUS) without detecting a causal variant. As such, compared to other testing methods, WGS often requires more time, increased data storage capacity, and a qualified bioinformatician to perform expert data analysis, increasing costs.

ES is faster and less expensive than WGS, but since it only involves sequencing the protein-coding regions of the genome, it comes with the risk of missing variants in noncoding regions. ES is also not optimal for detecting larger DNA structural variations, trinucleotide repeat expansions, and methylation anomalies.

Innovating tried-and-true chromosomal microarray analysis technology

Though chromosomal microarray analysis (CMA) has been around longer than NGS-based methods, the technology continues to evolve to meet the needs of today’s labs. CMA enables labs to investigate a broad range of genetic targets using a very small amount of sample to identify copy number variants (CNV).

Today’s arrays have improved upon the tried-and-true technology researchers have been using for years by minimizing sample requirements, enabling high-resolution copy number analysis, reducing inconclusive results and, perhaps most importantly, accelerating time to results. With faster turnaround times, labs can increase testing throughput with the same installed base to become more efficient and profitable.

Accelerating genetic analysis

While AI-enabled automation promises to accelerate the interpretation of cytogenetic research, clinical research labs should also look for innovations in the technology powering the front end of their genetic research studies. We are working closely with our research partners to understand their challenges and develop a new microarray solution with an accelerated workflow to transform cytogenetic research. Our vision is to take week-long processes and reduce them to a couple days without the need for large instruments, expensive computation equipment, or complex bioinformatics.