New algorithm could improve diagnosis of rare diseases

August 17, 2018 by Erin Digitale, Stanford University Medical Center
Credit: CC0 Public Domain

Today, diagnosing rare genetic diseases requires a slow process of educated guesswork. Gill Bejerano, Ph.D., associate professor of developmental biology and of computer science at Stanford, is working to speed it up.

In a paper published July 12 in Genetics in Medicine, Bejerano and his colleagues describe an algorithm they've developed that automates the most labor-intensive part of genetic diagnosis: that of matching a patient's genetic sequence and symptoms to a disease described in the . Without computer help, this match-up process takes 20-40 hours per patient: The expert looks at a list of around 100 of the patient's suspicious-looking mutations, makes an educated guess about which one might cause disease, checks the scientific literature, then moves on to the next one.

The algorithm developed by Bejerano's team cuts the time needed by 90 percent.

"Clinicians' time is expensive; computer time is cheap," said Bejerano, who worked with experts in computer science and pediatrics to develop the new technique. "If I'm a busy clinician, before I even open a patient's case, the needs to have done all it can to make my life easier."

A Phrank approach

The algorithm's name, Phrank—a mashup of "phenotype" and "rank"—hints at how it works: Phrank compares a patient's symptoms and gene data to a knowledge base of medical literature, generating a ranked list of which are most likely to be responsible for the symptoms. The clinician has a logical starting point for making a diagnosis, which can be confirmed with one to four hours of effort per case instead of 20-40 hours.

The mathematical workings of Phrank aren't tied to a specific database, a first for this type of algorithm. This makes it much more flexible to use.

Phrank also dramatically outperforms earlier algorithms that have tried to do the same thing, according to the paper. Bejerano's team validated Phrank on medical and genetic data from 169 patients, an important advance over earlier studies in the field. Prior studies had tested algorithms on made-up patients instead because real-patient data for this research is hard to come by.

"The problem is that this test [using synthetic patients] is just too easy," Bejerano said. "Real patients don't look exactly like a textbook description." On data from real patients, one older algorithm ranked the patient's true diagnosis 33rd, on average, on the list of potential diagnoses it generated; Phrank, on average, ranked the true diagnosis fourth.

Phrank also holds potential for helping doctors identify new genetic diseases, Bejerano said. For example, if a patient's symptoms can't be matched to any known human diseases, the could check for clues in a broader . "You might get the result that mouse experiments cause phenotypes similar to your patient, that you may have found the first human patient that suffers from this disease," Bejerano said.

Ultimately, "nobody is going to replace a clinician making a diagnosis," he said. But new technology could help experts use their time more efficiently, helping many more patients get diagnosed, he said.

Explore further: Automating genetic analysis helps keep up with rapid discovery of new diseases, study finds

More information: Karthik A. Jagadeesh et al. Phrank measures phenotype sets similarity to greatly improve Mendelian diagnostic disease prioritization, Genetics in Medicine (2018). DOI: 10.1038/s41436-018-0072-y

Related Stories

Automating genetic analysis helps keep up with rapid discovery of new diseases, study finds

August 15, 2016
When Shayla Haddock was born in 1997, her parents immediately realized something was wrong. The sixth of seven children, Shayla had unusual facial features. She had club feet and shorter-than-normal limbs. She was smaller ...

Genome analysis with near-complete privacy possible, say researchers

August 17, 2017
It is now possible to scour complete human genomes for the presence of disease-associated genes without revealing any genetic information not directly associated with the inquiry, say Stanford University researchers.

Individualizing health care one byte at a time

April 18, 2017
Genetic diagnosis of disease and personalization of treatment have the potential to dramatically improve strategies for diagnosis and therapy. Around 80% or rare diseases are thought to have a genetic component, but currently ...

A computer platform can pinpoint the genes behind rare diseases that have eluded diagnosis

December 17, 2014
A computer program that cross-references disease symptoms with DNA sequencing data can detect the faulty genes responsible for rare disorders with greater accuracy than other methods. Developed by scientists at A*STAR, the ...

Machine learning helps to predict the treatment outcomes of schizophrenia

July 12, 2018
Could the diagnosis and treatment of mental health disorders one day be aided through the help of machine learning? New research from the University of Alberta is bringing us closer to that future through a study published ...

Recommended for you

Why some human genes are more popular with researchers than others

September 18, 2018
Historical bias is a key reason why biomedical researchers continue to study the same 10 percent of all human genes while ignoring many genes known to play roles in disease, according to a study publishing September 18 in ...

Class of neurological disorders share 3-D genome folding pattern, study finds

September 18, 2018
In a class of roughly 30 neurological disorders that includes ALS, Huntington's Disease and Fragile X Syndrome, the relevant mutant gene features sections of repeating base pair sequences known as short tandem repeats, or ...

Researchers resolve decades-old mystery about the most commonly mutated gene in cancer

September 18, 2018
The most commonly mutated gene in cancer has tantalized scientists for decades about the message of its mutations. Although mutations can occur at more than 1,100 sites within the TP53 gene, they arise with greatest frequency ...

Clinical gene discovery program solves 30 medical mysteries

September 17, 2018
A table in a recently published paper tells the story of 30 families who have, sometimes after years of searching, finally received an answer about the condition that has plagued one or more family members. The Brigham Genomic ...

Study of one million people leads to world's biggest advance in blood pressure genetics

September 17, 2018
Over 500 new gene regions that influence people's blood pressure have been discovered in the largest global genetic study of blood pressure to date, led by Queen Mary University of London and Imperial College London.

Genetic mutations thwart scientific efforts to fully predict our future

September 17, 2018
Ever since the decoding of the human genome in 2003, genetic research has been focused heavily on understanding genes so that they could be read like tea leaves to predict an individual's future and, perhaps, help them stave ...

0 comments

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.