An ambitious project to create the most detailed picture of human genetic variation yet has been launched by British and American scientists.

The 1000 Genomes Project will map the tiny fraction of genetic material that differs between people, giving scientists unprecedented insight into why some are more susceptible to disease than others.

Scientists know that any two humans are more than 99% identical on the genetic level. The remaining fraction helps explain differences in people's health.

The $50m (£26m) map, which is expected to take three years to complete, will be made freely available, allowing scientists to pinpoint the genetic causes of common disorders swiftly and help tailor medical treatments to individual patients. Researchers will scan the genetic make-up of at least 1,000 people who either live in, or have ancestry tracing back to, Europe, Africa, Japan, China, India and Mexico. From these scientists will create a catalogue of genetic variation found in as few as 1% of the population.

"Such a project would have been unthinkable only two years ago," said Dr Richard Durbin, a geneticist at the Welcome Trust's Sanger Institute in Cambridge, who co-chairs the project. "We are moving forward to building a tool that will greatly expand and further accelerate efforts to find more of the genetic factors involved in human health and disease."

Since the 13-year-long Human Genome Project was completed in 2003, scientists have strived to look deeper into 3bn pairs of letters that form the human book of life. One, the HapMap project, has revealed large-scale genetic variations between populations, which have helped scientists identify more than 100 regions linked to conditions such as obesity, diabetes, heart disease and cancer.

Scientists look for genetic causes of disease by comparing genome scans from healthy people with sick people. But these studies usually only point to vague regions in the genetic code that may be hundreds of thousands of letters long.

"This project will give scientists a complete map of all the variation in that region, so they'll be able to use it to tease apart what might cause the illness and that will lead to better diagnosis and better understanding of the disease," said Durbin.

The project will begin this year with three pilot studies. The first will involve an in-depth scan of two families; the second will sequence the genomes of 180 people, while the third will examine 1,000 genes in 1,000 people. The main project will begin next year.