Kerstin Lindblad-Toh of the Broad Institute of Harvard and Massachusetts Institute of Technology in Cambridge, Massachusetts, has revealed that the genes were identified when her team was mapping the dog genome.
She claims that she and her colleagues are the first to achieve success in mapping the dog genome.
During the study, the researchers exploited the unique evolutionary history of dogs, which humans tamed from grey wolves between 15,000 and 100,000 years ago.
They have revealed that humans have bred dogs over centuries, selecting for traits like size and ability to heard sheep. Most of the 400 breeds of dogs descend from just a handful of hounds, they add.
The vast stretches of genetic similarity in dogs of the same breed enabled the research team to spot the few differences relatively easily.
"Here you have the perfect genetic model," Leif Andersson, a biologist at the Swedish University of Agricultural Sciences, in Uppsala, Sweden, who co-leads the project, was quoted by Nature magazine as saying.
The researchers used a gene chip that analysed about 27,000 single-letter differences across the entire dog genome, and thereby identified a region of 750,000 base pairs in the Rhodesian ridgeback genome that differed between ridged and unridged animals.
Thereafter, the researchers turned to Thai ridgebacks, which exhibit the same feature without being closely related to the Rhodesians.
Upon comparison of the same DNA stretch between the breeds, it was found that the real culprits for the prevalence of dermoid sinus in the dogs were the extra copies of four genes involved in foetal development.
The researchers have revealed that dogs lacking the duplication of genes are unridged, while the ones with one copy have normal ridge.
They also say that having two copies also carries an 80 per cent risk of dermoid sinus.
The mechanism paves the way for geneticists to use the dog genome to help identify genes involved in disorders such as diabetes that also affect humans.