What Were the Goals of the Human Genome Project?

The Human Genome Project, launched in 1990 by the US Department of Energy and the National Institutes of Health with international partners, set out to read the complete set of human DNA instructions for the first time and turn it into a tool for understanding disease.¹ Its goals went beyond reading DNA, covering new technology, other species, open data and the ethical questions genetic knowledge would raise.² It finished ahead of schedule, though its “complete” genome still had gaps that took two more decades to close.³⁴

The Official Goals Set in 1990

The founding plan boiled down to four goals:

Map the human genome and determine the order of all 3.2 billion DNA letters.⁵
Map and sequence the genomes of other organisms important to biology.⁶
Develop new technology for analyzing DNA.⁷
Study the ethical, legal and social implications of genome research.⁸

A genome is the full set of DNA instructions found in nearly every cell of a body. DNA is written with four chemical letters, often shown as A, C, G and T, and each position in the code is called a base pair. Reading the genome means recording the order of those letters. That order works like spelling: it determines which proteins a cell can make.

In 1990 the DOE and NIH published a plan for the first five years of a project expected to take 15 years.⁹ The underlying purpose was to give scientists tools to find genes involved in both rare and common diseases.¹⁰ Educating the public about genetic technology was named as a priority alongside the science.¹¹ Goals were set in five-year chunks because nobody could predict what technology would exist a decade later, and the goals were revised in 1993 for the 1993 to 1998 period.

Mapping Before Reading the DNA

A genetic map shows which markers, recognizable spots in the DNA that vary between people, tend to be inherited together. Early genetic maps used over 550 markers, enough that over 95% of the genome was linked to at least one.¹² Later, a high-density map with markers packed closely along every chromosome met one of the project’s first goals.¹³ With such a map, researchers could trace which stretch of DNA travelled with a disease through families.

That is how a map helps find a disease gene. When a marker is reliably inherited along with a condition in affected families, the gene is likely near that marker on the chromosome. The search across billions of letters shrinks to a much smaller region that can be examined in detail. The closer a marker sits to the gene, the less often the two are separated when parents pass DNA to children, which is what makes the clue useful.

A physical map is different: it pins landmarks to actual positions along the DNA molecule itself. Both were needed to line up millions of sequenced pieces. Sequencing couldn’t simply begin at one end because DNA could only be read in short pieces at a time. Without landmarks, those pieces couldn’t be placed in order.

A goal was a physical map of 30,000 landmark sites along the DNA by 1998.¹⁴ One map reached 15,086 landmarks, about one every 199,000 letters.¹⁵ Half the target was already enough to start sequencing 75% of the genome.¹⁶

Making DNA Reading Faster and Cheaper

The project sought sequencing at a cost well below 50 cents per DNA letter.¹⁷ It also wanted at least 10 times more DNA read than existing methods managed.¹⁸

Earlier methods ran DNA through flat gel slabs by hand, which was slow and needed a lot of skilled labour. Machines using many thin capillary tubes running in parallel, with bright fluorescent dyes, emerged from this push.¹⁹ DNA fragments travel through hair-thin tubes, and the final letter of each fragment carries a dye that glows a different colour under a laser. A detector records the sequence of colours as fragments pass, and software turns that into a string of A, C, G and T. Because many samples run at once and the process is automated, far less human labour goes into each letter read. Partly through this focus on technology, the project finished early.²⁰

The effects reach you directly. Cheaper sequencing made genetic testing in clinics practical, and consumer ancestry and health DNA kits rely on technology descended from this push. Prices kept falling after 2003 well beyond the original targets.

Why Other Species Were Part of the Plan

Human gene functions are often revealed by studying matching genes in other species.²¹ So the project sequenced E. coli bacteria, baker’s yeast, a roundworm, the fruit fly and the lab mouse.²² Comparing whole genomes gives insight into evolution and how bodies build and run themselves.²³

These animals were chosen because scientists already knew them well from decades of lab work. Their smaller genomes were also practice for the far larger human one. Mice share many disease genes with us, so they model human illness, and experiments impossible in people can be done in all of these species.

Genes with similar jobs in different species often keep recognizably similar spellings, because they were passed down from a common ancestor. If researchers find a human gene whose spelling matches a fly gene, they can look at what the fly gene does and form a good first guess about the human one. That guess is a starting point; it still has to be checked in human cells or patients.

Sharing the Data With Everyone

At a 1996 meeting in Bermuda the project agreed to release its DNA sequences into the public domain daily.²⁴ Every funded stretch of 1,000 letters or more went to free public databases in the US, Europe and Japan.²⁵ It was the first rule binding all international partners, not just US-funded labs.²⁶ Free tools such as the Ensembl website organized the sequence and gene predictions for anyone to browse.²⁷

Daily release was unusual: scientists typically kept data private until they published papers. Here, any lab worldwide could use results immediately, and the human sequence couldn’t be locked up as private property.

Public domain meant anyone, anywhere, could download the data without paying or asking permission. No restrictions prevented others from building on it, which kept it open to both academic and commercial users. People could still patent inventions made using the data; only the raw sequence itself was meant to stay open.

The Ethics Goal and Why It Was Built In

The founders made the ethics program a central part of the project from 1990.²⁸ The DOE and NIH spent 3% to 5% of their yearly project budgets on it.²⁹ That made it the world’s largest bioethics program at the time, and a model for others.³⁰ It became known as ELSI, short for ethical, legal and social implications.

Its goals included policy options on genetic testing services likely to be widely used.³¹ It also aimed to foster acceptance of human genetic variation and expand public and professional education.³²

The worries were real. People feared that employers or insurers could treat them differently based on their genes, and that the privacy of genetic results might not be protected. Families had to work out how to handle learning about inherited disease risk. There was also the risk of genetics being misused to rank groups of people. Rules and public understanding take years to develop, so starting in 1990 gave time to debate consent and privacy while testing was still rare.

Did the Project Meet Its Goals

The project exceeded its initial goals by 2003, two years ahead of its 2005 target.³³ Scientists from 20 institutions in six countries split the 23 chromosome pairs between centres.³⁴ Shared standards and the daily release of data let the pieces fit together into one assembled sequence. Five main centres in the UK and US sequenced most of the genome,³⁵ since large centres with many machines could process far more DNA than a small lab.

The company Celera raced the public project using a different method.³⁶ The public project mapped large chunks first, then shredded and read each chunk. Celera shredded the whole genome at once and relied on computers to reassemble it, which was harder because of repeated DNA stretches. Both drafts covered about the same amount of the genome.³⁷ The public method’s assembly was higher quality early, and Celera took a year to catch up.³⁸

“Finished” in 2003 meant the gene-rich, readable parts at high accuracy. Some hard regions were knowingly left for later.

The Gaps That Took Until 2022 to Fill

The reference genome long covered only the more easily read regions, leaving dense repetitive regions unfinished.³⁹ A team in 2022 presented a complete 3.055 billion-letter human sequence filling the remaining 8%.⁴⁰ It added nearly 200 million letters and corrected errors in earlier versions.⁴¹ The Y chromosome was still not gapless in that release.⁴²

Those regions were hard because they contain long stretches of near-identical repeated DNA. Short reads can’t tell which repeat copy they came from. Newer machines that read much longer pieces finally solved this.

A reference genome is a standard sequence that scientists compare other people’s DNA against. Gaps or mistakes in it can cause errors when researchers interpret a patient’s DNA in those regions. Filling the gaps gave a more trustworthy standard for comparison.

How Many Genes It Found

The project began assuming humans had about 100,000 protein-making genes.⁴³ The first genome papers in 2001 reported about 31,000 and 26,588.⁴⁴ Later counts fell to about 22,000, well below 30,000.⁴⁵ The 2022 complete sequence added 99 more likely protein-making genes.⁴⁶

Counting genes is harder than it sounds. They are found partly by computer prediction, which can miss short genes or flag stretches that aren’t truly genes. Deciding where one gene ends and another begins isn’t always clear, so different teams can land on slightly different counts.

Fewer genes didn’t mean simpler people. One gene can produce several different protein versions, and much of human complexity comes from when and where genes switch on. Gene number alone doesn’t track how complex an organism is.

What the Goals Led to in Medicine and the Economy

A public map of 1.42 million single-letter differences between people was built to help find genes for diagnosis and treatment.⁴⁷ A later project described genomes of 1,092 people from 14 populations, cataloguing 38 million such differences.⁴⁸

Any two people’s DNA is nearly identical, but they differ at many individual positions where one letter is swapped for another. Most of these differences have no effect on health, but some change how a gene works and can raise or lower the chance of a disease. Cataloguing them lets researchers compare groups with and without a condition to see which differences show up more often.

You may meet this legacy in several places. Genetic tests for inherited cancer risk depend on the reference genome. Ancestry kits compare your DNA to variation maps from many populations. Some cancer drugs are matched to tumour DNA changes, and newborn and pregnancy screening use genetic knowledge built on the project.

US federal spending on the project was $3.8 billion from 1990 to 2003.⁴⁹ Genomic research generated an estimated $796 billion in economic activity by 2010.⁵⁰ By an estimate from a research firm, each federal dollar was linked to about $141 of activity.⁵¹

1
Genome.gov. 1990: Launch of the Human Genome Project ↩︎
2
Genome.gov. 1990: Launch of the Human Genome Project ↩︎
3
Genome.gov. Human Genome Project Fact Sheet ↩︎
4
Science. The complete sequence of a human genome ↩︎
5
Genome.gov. 1990: Launch of the Human Genome Project ↩︎
6
Genome.gov. 1990: Launch of the Human Genome Project ↩︎
7
Genome.gov. 1990: Launch of the Human Genome Project ↩︎
8
Genome.gov. 1990: Launch of the Human Genome Project ↩︎
9
Genome.gov. 1990: Launch of the Human Genome Project ↩︎
10
PubMed Central. The Human Genome Project ↩︎
11
PubMed Central. The Human Genome Project ↩︎
12
PubMed. Construction of a linkage map of the human genome, and its application to mapping genetic diseases ↩︎
13
PubMed. A comprehensive human linkage map with centimorgan density ↩︎
14
Whitehead Institute. Whitehead Human Genome Map Ushers in Final Phase of U.S. Human Genome Project ↩︎
15
PubMed. An STS-based map of the human genome ↩︎
16
Whitehead Institute. Whitehead Human Genome Map Ushers in Final Phase of U.S. Human Genome Project ↩︎
17
Genome.gov. Improved Electrophoretic DNA Sequencing Technology ↩︎
18
Genome.gov. Improved Electrophoretic DNA Sequencing Technology ↩︎
19
PubMed. The use of capillary electrophoresis for DNA polymorphism analysis ↩︎
20
Genome.gov. Human Genome Project Fact Sheet ↩︎
21
U.S. Department of Energy Human Genome Program. HGP Research Area: Comparative and Functional Genomics ↩︎
22
U.S. Department of Energy Human Genome Program. HGP Research Area: Comparative and Functional Genomics ↩︎
23
U.S. Department of Energy Human Genome Program. HGP Research Area: Comparative and Functional Genomics ↩︎
24
PubMed Central. The Bermuda Triangle: The Pragmatics, Policies, and Principles for Data Sharing in the History of the Human Genome Project ↩︎
25
PubMed Central. Sharing Data to Build a Medical Information Commons: From Bermuda to the Global Alliance ↩︎
26
PubMed Central. Sharing Data to Build a Medical Information Commons: From Bermuda to the Global Alliance ↩︎
27
PubMed. The Ensembl genome database project ↩︎
28
Genome.gov. Review of the Ethical, Legal and Social Implications Research Program and Related Activities (1990-1995) ↩︎
29
U.S. Department of Energy Human Genome Program. Ethical, Legal, and Social Issues ↩︎
30
U.S. Department of Energy Human Genome Program. Ethical, Legal, and Social Issues ↩︎
31
U.S. Department of Energy Human Genome Program. Revised 5-Year Research Goals of the U.S. Human Genome Project (1993-1998) ↩︎
32
U.S. Department of Energy Human Genome Program. Revised 5-Year Research Goals of the U.S. Human Genome Project (1993-1998) ↩︎
33
Genome.gov. Human Genome Project Fact Sheet ↩︎
34
yourgenome.org. Who was involved in the Human Genome Project? ↩︎
35
yourgenome.org. Who was involved in the Human Genome Project? ↩︎
36
PubMed Central. On the sequencing of the human genome ↩︎
37
PubMed Central. Whole-genome shotgun assembly and comparison of human genome assemblies ↩︎
38
PubMed. A comparative analysis of HGSC and Celera human genome assemblies and gene sets ↩︎
39
Science. The complete sequence of a human genome ↩︎
40
Science. The complete sequence of a human genome ↩︎
41
Science. The complete sequence of a human genome ↩︎
42
Science. The complete sequence of a human genome ↩︎
43
PubMed Central. Open questions: How many genes do we have? ↩︎
44
PubMed Central. Open questions: How many genes do we have? ↩︎
45
PubMed Central. Open questions: How many genes do we have? ↩︎
46
Science. The complete sequence of a human genome ↩︎
47
PubMed. A map of human genome sequence variation containing 1.42 million single nucleotide polymorphisms ↩︎
48
PubMed Central. An integrated map of genetic variation from 1,092 human genomes ↩︎
49
Genome.gov. Calculating the economic impact of the Human Genome Project ↩︎
50
Battelle Memorial Institute. Economic Impact of the Human Genome Project ↩︎
51
Genome.gov. Calculating the economic impact of the Human Genome Project ↩︎