Schizophrenia affects 1% of the population and is considered one of the most debilitating mental disorders by many researchers and medical professionals. Unfortunately, it is also one of the most misunderstood. Schizophrenia is commonly misdiagnosed, and the prevailing stigma can make it all the more difficult to identify and treat.
The jury is still out on what causes schizophrenia. Most researchers attribute it to genetics, brain chemistry, environment, trauma, substance use, or complications at birth. The disease may be caused by one factor or an amalgamation of many.
A new algorithm developed by researchers in the Waterland Lab at Baylor College of Medicine has identified epigenetic differences between schizophrenic and neurotypical people. The algorithm, which utilizes machine learning, has inspired hope in researchers of early diagnosis and intervention for people with schizophrenia.
The Waterland Lab team’s algorithm is called SPLS-DA. Using machine learning, the algorithm can identify epigenetic markers to distinguish between people diagnosed with schizophrenia and those who are not. Moreover, it does so with 80% accuracy.
Epigenetics is concerned with how heritable factors like behavior and environment inflict changes on a gene without changing the gene’s DNA. An epigenetic marker indicates whether a gene is turned on or off. The algorithm tests the DNA from blood samples, inCORSIVs, a region of the human genome. Specifically, it analyzes methylation, an epigenetic marker, in this region.
Current evidence suggests that epigenetic factors may play a role in the presence or absence of schizophrenia. Therefore, understanding the relationship between epigenetics and the emergence of a disorder like schizophrenia is essential to understand how diseases work.
Dr. Robert A. Waterland, professor of pediatrics at the USDA/ARS Children’s Nutrition Research Center at Baylor, says that “DNA methylation at CORSIVs is established very early in life.” He goes on to say that the result of the algorithm study “indicates that the epigenetic differences we identified between schizophrenia patients and healthy individuals were there before the disease was diagnosed, suggesting they may contribute to the condition.”
Tests on epigenetics’s effect on schizophrenia have been run numerous times, but the Baylor team’s research provides new insights into where a significant epigenetic marker is located.
Their innovative approach combining machine learning and medicine has created a template that may not only be able to detect schizophrenia early but other diseases as well.
The true marvel of the test is that it considers confounding factors that were problematic in previous studies. Because methylation patterns are affected by smoking, diet, stress, antipsychotic medication, and exposure to chemicals, an accurate machine-learning algorithm must rule out these factors. The algorithm does just that and with a high level of accuracy.
Waterland and his team’s research implies that AI of this sort may be able to use information coded within an individual’s DNA to correctly identify the presence of a disease well before its onset. This is a huge breakthrough for a disorder like schizophrenia that, left untreated, can worsen.
Waterland says, “We consider our study a proof of principle that focusing on CoRSIVs makes epigenetic epidemiology possible.”