Scientists say that they have developed a blood test that can detect an internal clock

Can a blood test tell the time that it is in your body? Scientists at Northwestern University think so.

In a study published Monday in the medical journal PNAS, researchers reveal that they invented a new computer algorithm called "TimeSignature" that uses blood sample data to predict the circadian rhythm of someone.

Circadian rhythms are a major area of ​​research in the medical field. Scientists believe that someone's internal clock is the key to determining how their body works. The clock can direct the release of hormones, dictate when a person feels fatigue and help control daily changes in body temperature. The disruption of a normal rhythm can be a factor in many diseases: insomnia, heart disease and even neurological disorders such as dementia have all been linked to a misaligned circadian rhythm.

Getty Images / EyeEm

"Before, we had no clinically feasible way to evaluate the clock in healthy people and people with diseases. Now, we can see if a disturbed clock is correlating with various diseases and, more importantly, whether it can predict who will get sick, "said study co-author Ravi Allada, professor of neurobiology at Northwestern University. .

TimeSignature was created using machine-learning methods, which looked at real-world data from hundreds of blood samples of about 25 people. Artificial intelligence software used these data and found models to identify the internal markers that best match a person's physiological time.

Doctors currently have no way to measure a patient's internal clock. Models have been proposed, but none can currently be used in a doctor's office or hospital, and some models have been tested only on animals. Bioclock, for example, a program that incorporates machine learning to predict internal time, has only been applied to data from mice. Molecular times, another algorithm, have shown poor results in humans. Others are impractical, requiring frequent blood tests throughout the day.

TimeSignature has been tested exclusively in humans and would require only two blood tests spaced 10 to 12 hours apart. In order to obtain a correction of the circadian rhythm of a person, the algorithm first studied a set of more than 7000 genes, examining their maximal expressions at different times of the day. From these genes, TimeSignature found that 40 corresponded better to a person's circadian rhythm. He was able to predict the time of day according to the strength of the expression of these genes, comparing it with the actual blood-taking time of the patient. In the tests of about 50 additional patients, TimeSignature outperformed all other internal clock models. After taking the blood test, the artificial intelligence program could predict what the circadian time of the patient was to be in the space of two hours.

If TimeSignature works as well as they think, doctors might know "what time it is" in someone's circadian rhythm and use it to examine the impact of a disturbed "clock" on various diseases . The algorithm could also be used to guide when drugs should be taken – they may be more effective at times.

"It's crucial to know what time it is in your body to get the most effective benefits. The best time to take the drug for high blood pressure or chemotherapy or radiation therapy may be different from anyone else, "said Dr. Phyllis Zee, co-author of the study and chief of sleep medicine in neurology.

TimeSignature will soon not be used in medical practices, but the algorithm marks a milestone in the field. The software is available free to researchers and Northwestern has filed a patent on the blood test itself, which measures the genes expressed at the time of blood collection.

In the future, if doctors can plan their treatment around the patient's internal clock, medical care may improve.

Dr. Jonathan Steinman is a physician and author in radiology with the ABC News Medical Unit.