The Genomics Future

Craig Venter "The Genomic Future"

#include <stdheader> These are my notes from Dr. Venter's talk at Keck Graduate Institute on 7 Nov 2002. All editorial comments are my own. Any mistakes are my own. Any misrepresentations of Dr. Venter's opinions are my error. Mea culpa.

HISTORICAL BACKGROUND (SUMMARY)

• 1984: Venter's lab effort to sequence the adrenalin receptor (1200 nt) at NIH. took dozens of workers and years of time. Thinks "There has to be a better way."
• 1990: beginnings of a Human Genome Project
[Editor's note: I'm going to acknowledge my alma mater, UC Santa Cruz http://www.ucsc.edu, as being a one of the sites for the start of glimmer of an inkling of what would become the Human Genome Project. It is now an important site for the analysis of the raw Celera data.
http://genomesymposium.ucsc.edu/fs-1985SCW.html
http://genomesymposium.ucsc.edu/fs-bioinformatics.html
http://www.ucsc.edu/currents/00-01/06-25/genome.html
• 1994: First "whole", non-viral genome: E coli (~1 Mb).
• 1998: Drosophila fruit fly genome; Annotation jamboree.
• 2000: Begin shotgun sequencing of human genome. Used knowledge developed from the Drosophila effort in 1998. 3 women, 2 men (unidentified and anonymous). Published simultaneously with the public (scaffold) effort in 2001; 283 co-authors (biologists, mathematicians, programmers, etc.)
• New kind of "big science"
• Threw big $$$ at academics
• Over half million lines of computer code
• Refutatation of the "peer review" system of funding (National Institutes of Health http://www.nih.gov) who said that automated, shotgun sequencing was impossible.
[Editor's note: actually only 95% completed with many holes and errors; incorrect assembly of the pieces (Venter blamed "public" data as faulty.)]
• 2002: "Now"
• Mouse genome: done in 3 months
• have 88 complete prokaryote (bacterial) genomes
• KEY BREAKTHROUGHS:
• Use of "paired ends" to create large, lattice structure for organizing the shotgun sequencing effort
• Computational methods to assemble shotguns
[Editor's note: A list of completed genomes at http://bmerc-www.bu.edu/information/all_genomes.shtml

LESSONS FROM THE HUMAN GENOME PROJECT

• public data focused on gene dense chromosomes 4, 19
• rest of chromosomes "gene deserts"
• need more sequences; true power/knowledge is in comparison
[Editor's note: still need a genomic Rosetta Stone!]

• Only 26-36K genes (much fewer than the 60-100k expected; may still be that high if include splice varients, etc.)
• Only ~6500 genes were "known"; 11,226 ortho-homologs, 8619 de novo+two experimental data points, 12731 de novo+one experimental evidence
• Large scale genome duplication (eg ch 18 ~ part of ch 20)
• Deconvolute evolutionary history of the human species
• Appears to be 5 whole genome duplications from the worm C. elegans to mammals

• So far, 150 human genes NOT in mouse
• Otherwise 95% match
• All told, humans are ~1.25% (sequence) different from chimpanzee.

WHAT IS THE GENOMICS FUTURE?

• Ability to sequence entire mammalian genome in seconds. Requires technological advances, especially computational power. Sequence entire human for $1000 (ie, "when taking a baby home from the hospital, parents get CD-ROM of baby's genome")
[Editor's note: why not make it an Apple iPod? [http://www.apple.com/pro/science/gilbert/]
• New kind of research: "genome-based hypotheses" depending on comparing genomes (for example, parasite (malaria), vector (mosquito), and host (human).
• So far, only about 50% of genes have "function" defined.
[Editor's note: I think this is an underappreciated fact. We have all the letters, but we have no idea what half of the words mean, and even when we know the words, we're just guessing.]
• Biological approach to the energy crisis [http://www.sciencemag.org/cgi/content/full/298/5599/1701]

  Comparisons across kingdoms of life (e.g. archaebacteria/ "extremophiles") to identify novel metabolisms. Eg, Methanococcus jannaschii, has ability to metabolize methane. Useful for converting waste gases into hydrogen via a "synthetic cell"

• "Fallacy of personalized medicine". It will NOT be about individualized drugs for each person. Instead, it will be use predictive toxicology (ie, are you more suspectible to smoke, fats, etc.?) to select more/less effective drugs while avoiding side-effects.
• May boil down to "race-based" medicine where certain ethnic groups have particular symptoms and thus selective drug regiments
• Must acknowledge genes correlation with disease; thus, only contributatory factors only (genes are not CAUSITIVE or DETERMINISTIC)
[Editor's note: really that should be "allelelic" correlation. To a first approximation, we all have the same genes, just slightly different copies ("alleles"). Think of it as having slightly different editions of the same book.]
• Doctors/insurance companies must seperate genetics vs. environmental causes of disease. Believes the Doctor/Patient relationship will be unchanged, however the patient will be empowered. Will demand a change in how insurance works (only works by distributing risk across the whole market--can't "isolate" the "uninsurable".
• Analysis showing genetic bottleneck during the Plague may have led to human alleles that are resistant to HIV infection of T-cells.
[Editor's note: see refs. http://www.dellarco.com/ontarget/bodies/genetics.html and http://www.science-frontiers.com/sf119/sf119p05.htm]
• The "terabyte" challenge: genomics will be the next "killer app" for computational advancement. Ability to compare 100's of genomes at a time. Even if only 30,000 genes, likely 300,000 proteins or protein varients to keep track of!

QUESTIONS AND ANSWERS:

• Q: A vocal advocate of the Hydrogen Economy is Jeremy Rifkin, a long-time opponent of genetic engineering. What do you think about going head-to-head with him on using a biotech approach to hydrogen fuel?

A: (humorous response) "Are we doing something wrong?"

(serious response)Does advocate a change in the power structure toward local production and avoiding transportation of oil (safer, less conflict). If it can be done with biotech, why not?

• Q: What do you think of whole population sequencing such as the project in Iceland (DeCODE) [http://www.sciencemag.org/cgi/content/full/278/5338/566]

A: advantage of such projects is creation of linkage maps (not a new technique) by virtue of long pedigrees and homogenous population. However, having thousands of genomes allows limited population statistics for identifying SNPs (single nucleotide polymorphisms). Dr. Venter draws the parallel that Iceland as a society has established a hydrothermal economy and taken significant steps toward developing an infrastructure seperate from the oil economy. They are doing the same with genetic information.

• Q: Is genomics exascerbating the gap between rich and poor countries?

A: Poor countries are having more and health problems like the rich countries. However, they are NOT being helped by the way Big Phama is run now. The early history of drug development and patenting had poorer countries "stealing" technology from richer countries. This needs to be continued to "give" developing nations technology and intellectual property instead of the rich demanding the poor "buy" products and services.

• Q: Does the intellectual property restrictions limit academic research? For example, Myriad Genetics's monopoly on use of the breast cancer genes, BRCA1/2?

A: No.

[Editor's note: I utterly disagree!
http://www.boston.com/globe/magazine/2002/0224_patent_part1.htm
http://www.gov.on.ca/health/english/news/speech/sp_091901_tc.html
• Q: In the race to finish first, the human genome project has been criticized as being of poor quality. Is there a quality problem?

A: No--but the only remedy is to do more sequencing to compare methods and results. There are likely many indels (insertion/deletion) and SNPs to be identified among people.

• Q: You critisized how the NIH refused to fund your proposal to shotgun sequence the whole human genome in 1 year. How would you fix science funding?

A: The government/NIH is too focused on "hypothesis-driven" research and not enough "discovery" or descriptive science (ie, Darwin exploring the Galapagos Islands).

[Editor's note: Venter doesn't really offer a solution.]
• Q: What are the limits of genomics? What about proteomics/metabolomics/etc?

A: It is only the beginning of genomics. Need more varient information for comparisons.