After developing new methods for analyzing structurally complex regions of the genome, researchers have gained new insights into the genetics of obesity.

Specifically, the new study exonerates an enzyme called amylase 1, which is present in saliva and converts complex carbohydrates to simple sugars. The amylase gene is in a part of the genome that appears to be under construction – copy numbers vary widely between individuals, with anywhere from two to 17 copies present. They also appear to differ between populations, with populations who consume a high-starch diet having more copies on the average.

And, Steven McCarroll told BioWorld Today, "it's not just copy number." The part of the genome where the amylase gene resides also features rearrangements and inversions.

Collectively, that structural complexity makes the genome region that contains amylase very hard to map precisely. And that is why previous studies had come to differing conclusions on whether the amylase 1 gene affected obesity risk.

Genomewide association studies had not found a link between amylase and body mass index (BMI). But a study that specifically looked at the relationship between how many copies of the amylase gene an individual possessed found a very strong protective effect of amylase – the more copies an individual had, the less likely they were to be obese.

McCarroll, who is at Harvard Medical School, and his colleagues were intrigued by the discrepancy between those studies. And so they set out to develop a better, what he termed, "playbook" to analyze complex genome regions. One component of that playbook, he said, is "the development of molecular methods for the really precise measurement of copy number."

By using whole genome sequencing in combination with bioinformatics methods the team developed, they were able to do such precise measurements. In two separate cohorts, such precise measurements showed no relationship between amylase copy numbers and BMI.

The team was also able to discern an underlying simplicity in the complexity of the amylase locus. The widely differing structures appear to result from combinations of only a few underlying structures.

Their results appeared in the June 22, 2015, online issue of Nature Genetics.

McCarroll noted that although the question of whether amylase contributes to obesity is an important one, the methods his team has developed are much more broadly applicable.

"Amylase is at the leading edge of a much larger problem," he said. "We believe there are many hundreds of regions in the genome" that are too complex to analyze with the tools that can give insight into simpler genomic areas.

"Most of what has been learned about the relationship between copy number variation and disease risk," he added, "has been learned from simple deletions or duplications." But "many [complex] loci have real potential to be functionally important."

One reason that complex loci are complex is that they continue to evolve to a greater extent than other regions of the genome, and that in itself is one clue that they might be under selective pressure because they are functionally important.

McCarroll was noncommittal about such a possibility.

"We'll see. It's easy to speculate about," he said. After-the-fact speculation about the evolutionary reasons for why things are as they are can come perilously close to Just-So Stories.

Nevertheless, "it's variation that looks important," he acknowledged. Whether such loci, individually and collectively, turn out to be particularly important functionally is an empirical question. "But it seems pretty compelling to evaluate them systematically."

His own team is looking at a number of such complex regions, and in some of those regions, he said, they are uncovering functional relevance of copy number variations.

And in the meantime, "it's exciting that we are starting to understand these parts of the genome that until recently have been black holes."