Most genetic searchers for the roots of autism have compared massive populations of those on the spectrum with huge numbers of unrelated peers who are not.

A new statistical model, however, has focused on “discordant siblings” – the comparison between the entire genomes of affected and unaffected siblings.

The Proceedings of the National Academy of Sciences study finds that autistic individuals were genetically more like other unrelated autistic people, than they were like their own siblings, in target locations.

The paper compared 16,500 genomes from about 4,000 families, the scientists from Cold Spring Harbor Laboratory and Albert Einstein College of Medicine and elsewhere. The discordant siblings in the group were collected in the Simons Simplex Collection, and contrasted with full genomes from the Autism Genetic Resource Exchange.

“Ancient variants,” which have been within human DNA for thousands of years, were found among those on the autism spectrum.

The statistical analysis shows that autism appears to be caused by a large series of these small genes, and not large devastating variants. Those large devastating genetic changes appear to be weeded out of the genome within a generation, since the individuals are less likely to have offspring, due to their condition, they report. But the small are passed through families over the longest timelines.

“These undiscovered variants would have persisted sufficiently long in the human population to become widely distributed but by their sheer number, have escaped individual identification,” they write.

But their statistical method of assessing variant SNP arrays can only go so far, the authors concede. Whole-genome sequencing might better get to the heart of the variants, and their accumulation, they add. But their look into the differences within families – and what siblings inherit from their two parents in their DNA – could be a fertile ground for future genetic discoveries, they add.

“There is more power in family studies than we actually know how to tap into at this point,” said Michael Wigler, lead author, and geneticist at Cold Spring Harbor. “There is more information in a family structure than in the isolated person who’s got a disorder. Certainly this is true when dealing with de novo or germline mutation, but true even when examining transmission, as we did in the current study.”