A meta-analysis has found that around a fifth of variation in body mass index (BMI) is due to common genetic variation.
The study revealed 56 regions of the genome that had previously not been linked to the disease and adds support to the idea that multiple genetic variants play a role in predisposing people to obesity.
'Our work clearly shows that predisposition to obesity and increased body mass index is not due to a single gene or genetic change,' said senior study author Professor Elizabeth Speliotes from the University of Michigan Health System.
'The large number of genes makes it less likely that one solution to beat obesity will work for everyone and opens the door to possible ways we could use genetic clues to help defeat obesity,' she added.
The research team, who published their findings in two papers in Nature, analysed data from more than 300,000 people.
In the first analysis, they found a total of 97 genetic loci associated with BMI. The researchers estimate that, together, these loci account for only 2.7 percent of variation in BMI. However, after factoring in the collective influence of single nucleotide polymorphisms (whose individual effects are too small to be picked up in genome-wide association studies), they calculate that as much as 21 percent of variation in BMI is down to genetic differences.
The team found that the genes seemed to be implicated in pathways in the central nervous system, including those involved in the regulation of feeding and fasting, as opposed to metabolic processes.
Professor Speliotes told The Guardian: 'Looking at obesity, we didn't necessarily expect to see genes that work in the brain. In retrospect it's not that surprising that appetite and feeding pathways have a big role.'
In a second paper, the team also looked at genes linked to body fat distribution, revealing 49 loci associated with waist-to-hip ratio (adjusted for BMI).
They found an overlap between these loci and regions of the genome implicated in metabolic disease, such as insulin resistance, cholesterol levels and fasting glucose levels
The findings, from over 200,000 people, also revealed significant differences between men and women in the genetic regulation of fat distribution. Indeed, of 20 loci that had a sexually dimorphic effect on waist-to-hip ratio, 19 had a stronger effect in women.
'We need to know these genetic locations because different fat depots pose different health risks,' said co-author Professor Karen Mohlke, from the University of North Carolina.
'If we can figure out which genes influence where fat is deposited, it could help us understand the biology that leads to various health conditions, such as insulin resistance/diabetes, metabolic syndrome, and heart disease.'