Scientists have created an accurate method to analyze metabolic biomarkers determining whether a child is on the autism spectrum, according to a study published in the journal PLOS Computational Biology. The algorithm, based on a blood sample, is the first physiological test for autism and can help diagnose patients earlier.
The autism spectrum disorder includes a cluster of neurological disorders defined by troubles with social communication and restrained repetitive behaviors. Individuals with autism also have one or more accompanying ailments, such as epilepsy, intellectual handicaps, language impediments and sleep disorders. Autism affects 1.5 percent of the population, with the Center for Disease Control estimating in 2014 that one in 68 children in the United States had the disorder. Boys are more likely to be diagnosed than girls.
Autism has a strong hereditary factor, but environmental influences have also recently emerged as vital backers to its etiology, or the study of causation, and its pathophysiology, the medical subject that describes the conditions of a disease. However, no accepted biomarker for a diagnosis of autism exists. While there are distinct differences in the pathways of the brain, researchers have trouble pinpointing a single measurement of these pathways that split children with autism from neurotypical control groups. This absence of biological knowledge restrains diagnoses to be made based on behavioral monitoring and psychometric tools.
While autism is currently diagnosed and combatted with psychometric tools, researchers emphasized a biochemical view on this study. The study involved children between the ages of 3 and 10. Of the test participants, 83 children had autism and 76 children did not have the disorder. After collecting blood samples from all 149 children, the researchers measured for 24 metabolites, or the products of metabolism, from each sample. Researchers then utilized the Fisher Discriminant Analysis, a proportion reduction tool that expands on the differences between numerous categories. After crossing out data from one child in the group, researchers applied the dataset to analysis methods and used the result to create an algorithm. The algorithm then made a prediction about the data from the eliminated participant. Researchers cross-validated the results, switched different children out of the group and repeated the process for all remaining children.
Previous researchers have examined individual metabolites and found potential correlations with autism, but the relationship has been unresolved. The researchers used more advanced strategies, revealing patterns that would not have been evident with past attempts. Most examinations only inspect one biomarker, one gene and one metabolite.
While there were some differences, these results could not be accurately reproduced. Juergen Hahn, head of the Rensselaer department of biomedical engineering, explained that using data methods that inspect a set of metabolites linked with autism make a more powerful case.
The Fisher Discriminant Analysis correctly recognized 96.1 percent of all neurotypical participants and 97.6 percent of the participants with autism. Siblings of children with autism were found to be strikingly more alike to their neurotypical peers than to their siblings, even though they are inherently closer to their siblings than children in the neurotypical control group.
More research is anticipated because it is uncertain whether the test’s initial achievement could be applicable to children younger than 3 years old. Hahn expressed that evaluating children 18 to 24 months of age is a primary goal, but researchers have not yet performed this study so they are not aware of possible barriers.
The test may also be able to predict the beginning of autism in children who have not yet developed any clinical symptoms of the disorder.
There are concerns about the methodology used. A variety of previous studies had data that hinted at an existence of a single or a mixture of molecular changes that could separate children with autism from those without autism. It was mentioned that it would be unexpected to see a single molecular diagnostic test that would suffice for all the multiple subtypes of autism