A research team at the University of Warwick claims to have developed blood and urine tests that can effectively indicate autism in children. The study, using complex machine learning algorithms designed to identify differences in blood and urine between autism spectrum disorder (ASD) subjects and healthy children, suggests a variety of new biomarkers that could be related to the condition.

A big hunt is currently underway for a clear biomarker that can allow doctors an objective pathological way to diagnose ASD. Currently the only way to diagnose the condition is through behavioral assessments, and most children aren't identified as autistic until after the age of four. The condition is undeniably complex, with many researchers recognizing the causes as rooted in an elusive combination of genetic variants and environmental factors.

Previous research has suggested an assortment of possible biomarkers that could offer physicians potential diagnostic tests, from analyzing single protein levels in saliva to complex combinations of metabolites in a blood sample.

The latest research in the area from the University of Warwick suggests a link between ASD and the presence of certain damaged proteins in blood plasma. The study claims that children with ASD had higher levels of the oxidation marker dityrosine, as well as what are called advanced glycation end products (AGE).

Much like similar work from the Rensselaer Polytechnic Institute in 2017, this study used machine learning techniques to study the blood of both children with and without ASD in order to distinguish chemical differences between the two groups. The study looked at blood and urine samples from 38 children diagnosed with ASD, compared to a healthy control group of 31 children. All the subjects studied were aged between five and 12.

Experts in the field are suggesting that while this is certainly an interesting study, it is in no way close to providing a new diagnostic test for autism. James Cusack, the Director of Science at UK charity Autistica, and not affiliated with the research, says this study is weakened by its small sample size and, "does not tell us how effectively this measure can differentiate between autism and other neurodevelopmental or mental health conditions such as ADHD and anxiety."

Max Davie from the Royal College of Paediatrics and Child Health, who did not work on this study, also questions the validity of the data, pointing out, "The analysis was derived from children whose ages averaged 7-8, so there is no data to indicate that very young children will have the same metabolic pattern and that the results found would be reproducible in infants."

The team behind the new study certainly recognize some of the limitations in the work, admitting that further research needs to be done to isolate different age-related biomarkers, and to identify variations in severity of the condition. While trying to uncover a clear pathology to this devastating and mysterious condition is undeniably important work, there are numerous questions that still remain unanswered, meaning we are likely still a long way from a test like this being broadly deployed.

"While we applaud the arrival of this interesting area of research, it is important that it is not adopted with too much enthusiasm – if applied to a large population it will produce large numbers of 'false positives', causing huge worry and potential harm to children and families," adds Davie.

The new study was published in the journal Molecular Autism.