Human Genome Project

The Human Genome Project was an international research program that produced the first sequence of the human genome. It ran from October 1, 1990, to April 14, 2003, and was carried out by the International Human Genome Sequencing Consortium, 20 universities and research centers in six countries, with United States funds from the National Institutes of Health (NIH) and the Department of Energy. Its leaders had planned for 15 years and finished in about 13. The sequence it published covered over 90% of the genome, and the last gaps in a human reference sequence were closed in 2022.^[1]

The project grew out of a question about mutations. In December 1984 the Department of Energy and an international commission on environmental mutagens held a meeting in Alta, Utah, to ask whether new DNA methods could detect any rise in mutations among the survivors of the atomic bombs; George Church was one of its 19 participants.^[2] A workshop at the University of California, Santa Cruz, followed in May 1985, and the Department of Energy held another in Santa Fe in March 1986 to judge whether sequencing the human genome was feasible. In April 1988 the National Research Council and the congressional Office of Technology Assessment each recommended a concerted program.^[3]

The NIH created an Office for Human Genome Research in 1988, which became the National Center for Human Genome Research on October 1, 1989, with James Watson, who had co-discovered the structure of DNA, as its first director. It was renamed the National Human Genome Research Institute in January 1997. In April 1990 the NIH and the Department of Energy published a plan for the first five years of a 15-year project whose goals included mapping the genome and "determining the sequence of all its 3.2 billion letters", mapping and sequencing the genomes of other organisms, and developing technology. The project began officially on October 1, 1990.^[3] An ethics program, the ELSI program, was set up in 1990, and Congress later required the institute to spend at least 5% of its research budget on it.^[1]

The project used Sanger sequencing, a method published by Frederick Sanger and colleagues in 1977 in which chain-terminating nucleotides stop copies of a DNA fragment at each base, so that the fragments can be sorted by length and read in order.^[4] Reads were short, so the DNA was first cut into large pieces and cloned in bacteria, the clones were mapped to their positions on the chromosomes, and each clone was then sequenced separately and assembled, an approach called hierarchical shotgun sequencing.^[5] The maps came first: a genetic linkage map in September 1994, a year ahead of schedule, and a physical map in December 1995. Chromosome 22, with 33.5 million letters, was the first human chromosome sequenced, in December 1999.^[3]

A rival effort used a different plan. Celera Genomics, a private company, broke the whole genome into random fragments and assembled them by computer without mapping them first. Its paper of February 2001 reports 14.8 billion base pairs of sequence read in nine months from the DNA of five people, and it combined those reads with the public data, which had been cut into segments of 550 base pairs.^[6]

At a meeting in Bermuda in February 1996 the project's leaders agreed that all human genomic sequence from federally funded large-scale centers "should be made freely available and placed in the public domain within 24 hours of being generated". The principles ran against the usual practice, which was to release experimental data only after publication, and the leaders reaffirmed them in Bermuda in 1997 and at Cold Spring Harbor in May 1999.^[3]

On June 26, 2000, the consortium announced a working draft at a White House ceremony. The draft sequence, published in February 2001, covered about 90% of the genome and had more than 150,000 gaps. It estimated the number of human genes at about 35,000, a figure later revised to about 20,000, and it showed that the DNA sequences of any two people are 99.9% identical.^[3] On April 14, 2003, the consortium announced the project complete, at 92% of the genome and with fewer than 400 gaps.^[1] The finished sequence, described on October 20, 2004, held 2.85 billion nucleotides with 341 gaps, covered about 99% of the euchromatic genome, and had an error rate near one in 100,000 bases. It put the number of protein-coding genes at 20,000 to 25,000.^[7]

The reference was a patchwork. About 70% of it came from one volunteer, and the rest from 19 others, all anonymous; most of the donors were recruited in Buffalo, New York, through newspaper advertisements, and their blood was prepared at Roswell Park Cancer Institute.^[1] On March 31, 2022, the Telomere-to-Telomere consortium announced the first complete sequence, which was 3.055 billion base pairs long and gapless for every chromosome but Y. It covered the remaining 8% of the genome and added nearly 200 million base pairs to the previous reference.^[8]

The projected cost was $3 billion. The United States spent about $2.7 billion across the project's wide range of activities. The National Human Genome Research Institute estimates that the draft sequence cost about $300 million worldwide over 15 months, from April 1999 to June 2000, with 50 to 60% from the NIH; finishing it cost about $150 million more.^[9] A second reference genome in 2003 would have cost about $50 million, a draft genome in 2006 about $14 million, and one in late 2015 less than $1,500.^[9]

The first of the faster machines came in 2005, when 454 Life Sciences described a sequencer that read DNA in picoliter reactors.^[10] Illumina and Ion Torrent followed with machines of their own.

In 2004 the institute began a program of grants for methods that would cut the cost of sequencing a human genome to $1,000 or less, and work on nanopore sequencing was among what it funded.^[11] A human genome was assembled from nanopore reads in 2018.^[12] The history is told in History of nanopore sequencing.

Church is a personal investor in Molecular Reality Corporation and was the first formal member of its scientific advisory board.^[13] In the autumn of 2009 the company's founder, Kent Kemmish, wrote a winning application for a Grand Opportunity grant from the National Human Genome Research Institute, made with Church's laboratory at Harvard Medical School, which was the formal awardee.^[14] On October 21, 2026, the company dedicated the design of the MR1 Nanopore Reader to the public domain.^[15]

• Halcyon Molecular
• DNA
• Nanopore sequencing