A team of researchers led by Baylor College of Medicine in the US has developed a new approach to potentially diagnose schizophrenia at an early stage by combining machine learning with a blood test.

Leveraging a machine learning algorithm referred to as SPLS-DA, this approach can assess certain regions of the human genome called CoRSIVs, to detect epigenetic markers for schizophrenia.

Epigenetic markers are a profile of methyl chemical groups in the deoxyribonucleic acid (DNA).

The epigenetic markers detected by the researchers from the DNA in blood samples varied between schizophrenia patients and individuals without the disease.

Based on the findings, the team created a model that can analyse a person’s chances of developing the condition. On testing the model with an independent dataset, it was able to detect schizophrenia patients with an accuracy of 80%.

Baylor College of Medicine paediatrics – nutrition professor Dr Robert Waterland said: “Schizophrenia is a devastating disease that affects about 1% of the world’s population.

“Although genetic and environmental components seem to be involved in the condition, current evidence only explains a small portion of cases, suggesting that other factors, such as epigenetic, also could be important.”

The researchers previously found a set of CoRSIVs in which DNA methylation, a common epigenetic marker, varies between individuals but is consistent across various tissues in one individual.

They expected that assessing CoRSIVs could be a new method to uncover epigenetic causes of disease.

Waterland added: “Because methylation patterns in CoRSIVs are the same in all the tissues of one individual, we can analyse them in a blood sample to infer epigenetic regulation on other parts of the body that are difficult to assess such as the brain.”

The latest study is vital as it analysed key potential confounding factors, the institute noted.

Factors such as smoking and the use of antipsychotic medicines, which are common in schizophrenia patients, can impact methylation patterns in the blood.

The researchers ruled out the effect of these factors on the methylation patterns, achieving stronger epigenetic signals related to the disease.