The massive DNA study, involving dozens of research centres worldwide, sheds important light on the biological basis of lung cancer and will help shape new strategies for treatment, the authors said.
"This view of the lung cancer genome is unprecedented, both in its breadth and depth," said Mathew Meyerson of Harvard and MIT, who led the research.
"It lays an essential foundation, and has already pinpointed an important gene that controls the growth of lung cells."
Each year some 1.3 million people die from lung cancer, making it the most lethal form of the disease, according to the World Health Organisation.
The new study focuses on lung adenocarcinoma, which accounts for just under a third of all lung cancer cases.
Part of the international Tumour Sequencing Project, the study looked for abnormalities in the DNA of more than 500 tumours from lung cancer patients.
Most human cancers stem mainly from DNA changes that accumulate in cells through a person's life, but the nature of these changes -- and their consequences -- has remained largely unknown.
The scientists used cutting-edge technologies to scan the human genome for markers called single nucleotide polymorphisms (SNPs) that highlight missing or duplicate sections of genetic code.
The study, published online by the British journal Nature, uncovered a total of 57 genomic changes that occur frequently in cancer patients.
Of these, at least 40 are associated with genes not previously known to be involved in lung adenocarcinoma.
The genetic anomaly that turned up the most frequently incriminates a gene called NKX2.1 as an accelerator of cancer cell growth.
NKX2.1 normally acts as a "master regulator" that controls the activity of other genes in cells lining tiny air sacs in lungs called alveoli.
The discovery that a gene functioning in a particular group of cells can promote cancer growth could help scientists design drugs to fight not just lung cancer but a wide range of cancers, the researchers said.
In addition, the use of powerful tools and technologies to sequence the genomes of lung cancer patients "represents a general approach that can and should be used to analyse all types of cancer," said co-author Eric Lander, director of the Broad Institute of MIT and Harvard.
The collaborative research behind the study has laid the groundwork for even more ambitious genome projects such as the Cancer Genome Atlas, which seeks to map the common genomic changes in a wide range of human cancers.
In its pilot phase, the Atlas project is focusing on the most common form of brain cancer, glioblastoma multiforme, as well as ovarian and squamous cell long cancer.