Berkley Synthetic Biology Class, April 5, 2006

1. Berkley Synthetic Biology Class, April 5, 2006 Commercial Gene Synthesis Technology John Mulligan

2. Topics • Commercial gene synthesis today • Issues and technology for the future • Governance and the potential for nefarious applications of synthetic biology

3. Access to DNA is Central to Modern Biology • Biomedical Research • Biology • Agriculture • New areas such as Synthetic Biology

4. Acquiring and Modifying DNA is Costly • Researchers spend > $800MM/year on reagents to clone and modify genes • Deutsche Banc Alex Brown, 2000 • Every $1 spent on reagents represents an additional $2 to $5 of fully loaded costs • Labor, overhead, facilities, etc. • Billions of dollars in time and effort every year • Roughly $1 billion in direct costs to NIH • $1-$2 billion to industry

5. Gene Synthesis Improves Research Productivity • Less costly than other methods for many projects today ($1.25 to $1.60 per base pair today) • Industrial groups believe their internal costs with other methods to be ~$2 per bp • Academic groups have lower costs but still find synthesis economical for many projects • Cost of synthesis continues to decline rapidly • Complete control of sequence allows improved experimental design and new experimental approaches • Use the perfect gene for your experiment instead of the gene you have in the freezer

6. Commercial Gene Synthesis • Potentially a substitute for $1 to $2 billion in fully loaded costs • We estimate the current market is $20 million to $30 million a year • Revenues rowing at 30% to 50% a year • Volume growth much higher • Highly fragmented: 50 or more companies in this area world wide • Still a tiny fraction of the overall molecular biology market • We expect it to grow rapidly but to take 5-10 years to reach a significant fraction of the molecular biology market

7. Blue Heron Order Mix • Standard Orders • $500 to $50,000 • One to ~50 genes, as fast as possible • Standard delivery schedule • High-Volume, Time-Sensitive (Corporate) • Hundreds of kilobases, as fast as possible • Negotiated delivery schedule • A 200 kb project in 2004 and a 450 kb project in 2005 • High-Volume, Time-Insensitive (Government) • Thousands of kilobases • Extended Delivery, discounted price • Enables full capacity utilization to leverage fixed costs and maximize economies of scale

8. Gene Synthesis Technology • In use since the late ’70’s but only beginning to be widely used • Challenges • Error rate: ~1/300 • Mismatched hybridization can lead to scrambled order • Reliability impacts speed and cost • Three general approaches • “One pot” ligation and/or PCR • Convergent assembly • Solid phase assembly

9. PCR Amplification PCR Assembly Multiple oligonucleotides in a single reaction.

10. PCR Assembly • Simple to do, often works • The technology used by nearly all commercial providers • Many published protocols • Limitations • Some sequences are difficult or impossible to PCR • Difficult sequences can add to the cost and delivery time

11. Convergent Assembly A series ligation and purification steps, each involving only two fragments.

12. Convergent Assembly • A series of simple, reliable reactions • Works on almost any sequence • But, it is slow and more expensive than PCR-based methods for many genes

13. Inside column Solid-Phase Assembly Within each column, double-stranded oligos (duplexes) are sequentially added to a solid support, with intervening wash steps.

14. Inside column Solid-Phase Assembly Attach duplex to solid phase support

15. Inside column Solid-Phase Assembly Wash

16. Inside column Solid-Phase Assembly Add the second duplex into column

17. Inside column Solid-Phase Assembly Attach the second duplex to the first duplex

18. Inside column Solid-Phase Assembly Wash

19. Inside column Solid-Phase Assembly Add the third duplex into column

20. Solid-Phase Assembly Attach the third duplex Inside column

21. Inside column Solid-Phase Assembly Repeat to assemble complete fragments and elute from column

22. Solid-Phase Assembly • Simple reaction: two fragments, three ends • Drive reaction with molar excess • Wash away side reactions • Fully automated at Blue Heron

23. Oligos Cloned Sequence verified Restricted Subtarget Subtargets Final Clone Solid Phase Assembly of Whole Genes

24. GeneMaker® Overview • Proprietary software designs build strategy • Oracle database instructs instruments to build • Oligos are synthesized and hybridized • Patent-pending automated solid phase assembly • Cloning and sequencing • Error removal methods throughout process

25. Issues and Technology for the Future

26. Gene Synthesis is Complex • Every order is different • Every gene is made from a dozen to several thousand parts • Every part is new and used for only one order • The smallest parts are chemicals • Mixed populations of good and bad parts • Error rate of on in a few hundred • Larger parts are biological • Unpredictable behavior • The final product must be perfect

27. Existing Manufacturing Tools are Inadequate • Commodity market • Prices drop 30% to 50% / year • We must drop production costs at least this fast • Mass customization used in some industries • Have not found one where every part is new • Handling high failure rates is critical to controlling manufacturing costs • Existing tools focused on assembly-line production, “job shops”, custom engineering • None

28. Automated Laboratory vs. Manufacturing • Most or all gene synthesis today is carried out in sophisticated laboratories with some automation • PhDs involved • Difficult to scale rapidly • Within a few years, nearly all commercial gene synthesis will be carried out in manufacturing facilities • Largely automated • Robots for production • People for process development • Highly sophisticated process control and scheduling • Interesting, meaty problems for operations research…

29. Blue Heron is a Software Company • Integrated manufacturing system • Automated storage • Integrated materials handling: e.g., robot arm on a rail • Off the shelf components: pipettors and incubators • Proprietary process • Lots of software • Automated design of manufacturing process • Database control to track every fragment and manage rework cycles • Sophisticated scheduling • Integration software • Protocol software on individual instruments

30. Nefarious Applications and Governance

31. The Potential for Biowarfare Applications • Many researchers synthesize or clone pathogenic DNA as they work to understand the basic biology of the pathogen and to develop new therapeutics • Most viral genomes are within the range of today’s technology • Blue Heron delivered the fragments for a >25 kb virus in 2004 • Vaccinia is 180 kb • ould be done in 6-12 months, 40 SNPs • One or more bacterial genomes will be synthesized within the next year • Nefarious uses of synthesis are possible

32. Gene Synthesis Technology is Widespread Bioneer Corporation 49-3, Munpyeong-dong, Daedeok-gu, Daejeon 306-220, Korea “The capacity of this facility is to produce 7.2 tons of phosphoramidite per year… Currently we have (the) capacity of producing 20,000 oligos per day… Bioneer offers a special gene synthesis service.” But the vast majority of the sophisticated molecular biology capacity is in Europe and North America

33. Controlling Synthesis Technology is Difficult • Synthesis materials are easy to acquire • Any sophisticated chemistry group could build oligonucleotide synthesis capacity from scratch • For large-scale synthesis groups the “drop at the bottom of a reagent bottle” can add up to kilograms of phosphoramidite per year- tracking the materials is not feasible • PCR-based synthesis works on many sequences • Transforming and growing bacteria is low-tech

34. New Methods Extend Synthesis Capabilities Build genes with a modified ink-jet printer?

35. Garage Technology in Five Years? • Lone hackers with few resources NO • Governments or organizations YES • Any country or moderately well-funded group could put together the capacity FROM SCRATCH with a moderate investment ($500K and 3-6 PhDs)

36. Group “BW Hacking” • Technology access is easy • Robust, world-wide market for used equipment • Simple hardware for all aspects of the technology- could be built from scratch by a few engineers • Chemistry is feasible for companies or laboratories in many (nearly all?) countries • Molecular biology and bacteriology kits available from many different companies in many countries • Protocols on the internet • But, it is still far harder than organism- or tissue culture-based BW hacking • ~$1M, 3-6 key technologists, and a modest industrial infrastructure required for synthetic biology

37. Governance: Select Agent Regulations • Screen all orders against a database of select agent genes • Black Watch, Craic Computing • Review sequences that are similar to those genes • A Ph.D. reviews several positive hits per day • Most hits are not select agent genes • Detailed analysis of select agent genes • Check the literature • Discuss with customer • Decide if we can build the sequence

38. Current Regulations Require Interpretation • Many genes from select agents are not dangerous and are not controlled • E.g., bacterial metabolic genes • Many select agent genes resemble harmless genes • E.g., non-pathogenic relatives • Many scientists use non-functional parts of select agent genes in their research • Viral coat proteins for vaccine development • Enzymes for testing anti-microbial and anti-viral drugs • DNA fragments or proteins for development of diagnostics

39. Regulatory Clarity is Needed • Goals • Restrain/monitor access to dangerous DNA fragments • Retain ability to carry out rapid biomedical and other life science R&D • However, no national regulatory scheme can completely block the arrival of new pathogens • Moreover, poorly conceived regulation could impede ability to respond to new pathogens

40. Our Perspective on Regulations • Regulation should define the DNA sequences that are covered • Current select agent rules require interpretation • And action to be taken when regulated sequences are requested • What needs to be reported? To whom? What is the involvement of our customer in the process? • Regulations could shift the development our industry • If regulations require disclosure of all sequence orders, pharmaceutical researchers will not outsource gene synthesis because sequence data is confidential • Such regulation would lead to an instrument (“gene synthesis in a box”) market • The development and dispersion of such instruments would make the technology harder to control

41. Solution: Select DNA Sequence Database • A list of Select DNA Sequences • DNA sequences that could be used to build pathogens or to enhance pathogenicity • Actively maintained by an oversight panel and a set of organism-specific experts • Updated on a regular basis (e.g., monthly) • Select sequences defined in terms of a reference sequence and a percentage identity to the reference sequence • Current method of BLAST search against BlackWatch database results in many false positive “hits”, each requires time to research and identify risk

42. Select Sequences • Three classes of sequences • Select Agent Genes Require a permit • Related Genes Require reporting • All other genes No reporting required • Control of high-threat sequence • Tracking of sequences that could be incorporated into new pathogens • Fragments of select agent genes • Other pathogenic genes • Other sequences? • No reporting requirement for most sequences

43. Operational Considerations • Positive requirement to check orders against the Select Sequence database • Current rules make it illegal to provide certain sequences but do not require providers to check for those sequences • Clear procedures for identifying organizations and individuals that are authorized to possess molecules encoding Select Sequences • Centralized database to collate information on reportable sequences • One could now buy parts of a virus from several different providers and not violate any regulations until they were assembled

44. Gene Synthesis is an International Industry • Researchers are located all over the world • Gene synthesis companies exist all over the world • Dozen + in US • Dozen in Europe • Several in Asia (at least) • Ad hoc (non-commercial) gene synthesis occurs regularly in labs all over the world • US regulations cannot block nefarious access to this technology • US regulations can impact the efficiency of our response to pathogens

45. Rapid, Effective R&D is the Solution • Our response to new pathogens depends on decades of basic research AND the immediate application of today’s best technology • Gene synthesis could play an important role in rapid responses to new diseases • Scientists working for the good of society have an extremely large advantage in resources • We need to maintain and improve our R&D capacity to respond the this threat • Modest investments in current technology could reduce the danger

46. Summary • Gene synthesis and molecular biology are central to modern biological research • The technology is ubiquitous and international, thus control from within the USA is not possible • Current regulations need improvement • Clear definition of Select Sequences • Tracking of related sequences • Poor regulatory choices today could significantly reduce our ability to respond to new pandemics, whether natural or man-made