Bioinformatics

Bioinformatics is the marriage of genetics, mathematics, and computer science. Genetic information – DNA sequences of genes and components – is manipulated using mathematical models and computer programs. The field has developed in the past decade as the volume of genetic information evolved beyond our understanding of its significance, i.e. DNA sequences became known without knowing their function, expression, or significance to the cell and organism. The genetic information, as DNA sequences become increasingly known, thus serves as raw data and the foundation for the field in which tools are used to lead to understanding its significance.[1]

Bioinformatics aims to assist science in its understanding of the role of each component of DNA in the cell. While a variety of organisms are being studied,[2] human genomics have received the lion's share of research and media attention.[3] Beyond its significant implications to the advancement of science, bioinformatics has tremendous implications to commerce. A wider understanding of the significance of genetics promises to revolutionize the health care and health services industries through the development of techniques to combat ailments and improve quality of life. The software industry stands to gain from the trade in the tools used in bioinformatics. Perhaps the most potentially lucrative area of application is in the ability to investigate the genetics of disorders offering the opportunity for large pharmaceutical companies to search for and develop highly profitable products more efficiently.[4]

The commercial side of bioinformatics is multifaceted. The first step is the collection of genetic information in the laboratory from a variety of samples, and with a variety of aims and techniques. This sequencing stage is more closely aligned with scientific advancement, and results in the accumulation of raw data from which are drawn conclusions leading to further understanding of processes and principles. Second, the sequence data is manipulated using computer programs and mathematical models, typically in a search for genes or regions whose understanding and manipulation could lead to profit.[5] This stage is often repeated, combining previous manipulations in new models, further refining previously screened samples, and addressing results within the context of larger modelled systems. At this stage, research is conducted using highly complex and specialized computer software and mathematical modelling systems. Finally, the results of these manipulations are developed into marketable products and tested. For example, in the pharmaceutical industry, results of bioinformatic testing would lead typically to an area for potential chemical effectiveness, candidates would be identified and developed, and they would be tested for hazards and effectiveness. This technique represents a significant departure from previously known drug development: rather than observe the effect of various chemicals on the system and market the ones that are effective for specific purposes, with bioinformatics the genes and their defects (or functions) can be known in advance and candidates for their manipulation and interaction need only be developed or found. Thus bioinformatics presents a more streamlined approach to drug development, increasing the efficiency of the costliest component of the pharmaceutical industry, and represents potential for both greater accuracy, effectiveness, and range of products and increased profits.

The legal implications of bioinformatics are varied. A significant component of bioinformatics analysis uses specially designed software that is both highly complex and highly valuable.[6] Access to these (usually expensive) tools is governed by existing copyright and licensing laws. Privacy, especially as it concerns collection and distribution of genetic information, is an unresolved but as yet unrealized fear.[7]

The search for and collection of sequencing data is almost exclusively[8] conducted by publicly funded institutions.[9] As such, access to genetic information is rapid, free to academic researchers, and available for a nominal fee to commercial users.[10] For example, the human genome database, hosted by The Hospital for Sick Children in Toronto, is available on the internet and updated frequently as new developments warrant.[11] This culture of free access to the raw data of genetic information is motivated largely by the impetus in science of making information available for further scientific investigation and as the basis of credibility.[12] The activities of companies like Celera Genomics, however, threaten this culture by making access to the information a commercial venture. Privately funded institutions are more likely to be motivated by profit, and could impair scientific access to independently discovered or developed genetic information through expensive patents or licences. While this remains the fear of the scientific community, it has yet to be fully realized as the current culture is buttressed by strong public funding, the persistent impetus for free access to information, and public and academic support for cooperation in research.[13]

An as yet unresolved legal implication of bioinformatics is found in its unique nature of a marriage between biology, mathematics, and computers. While DNA is considered a patentable material, considerable controversy exists over the wisdom of this direction as DNA is also essentially a storage system of information.[14] Furthermore, even though patent prosecution is rarely initiated by the publicly funded institutions currently forming the foundation of bioinformatics research stations, the manipulation of that data has legal implications of its own. A debate exists over whether the computational manipulations of the genetic data constitute sufficient intervention and ingenuity to warrant the issue of a patent; the ethical wisdom and economic consequences of stifling investment are also in question.[15] In the current state of the field, however, most data gathering and sequencing is performed at academic and publicly funded institutions whose findings are readily and freely available as is required for credibility of conclusions and publication. Academic institutions that further use and manipulate that data likewise publish their findings as is customary in the scientific practice. Commercial institutions typically conduct their research using the available raw data to develop a marketable product, which is uncontrovertibly patentable under current schemes. As long as the scientific culture persists (or is protected by legislation) whereby researchers earn credibility in publishing their findings, the commercialization of genetic information seems unlikely in bioinformatics.

D. Banisar & S. Davies, "Global Trends In Privacy Protection: An International Survey Of Privacy, Data Protection, and Surveillance Laws and Developments" (1999) 18 J. Marshall J. Computer & Info. L. 1.

The privacy protection provisions of over fifty countries in the light of recent technological developments and the need for data protection legislation.

T. K. Baumann, "Proxy Consent And A National DNA Databank: An Unethical And Discriminatory Combination" 86 Iowa L. Rev. 667.

Privacy and consent issues are discussed with respect to the establishment of a national genetic database

T. Caulfield, "Underwhelmed: Hyperbole, Regulatory Policy, and the Genetic Revolution" (2000) 45 McGill L.J. 437-460.

The current debate and concerns surrounding the genetic revolution are blown out of proportion as industry has produced fewer than expected products and public debate centres around human cloning. Parliament should take a more balanced and reasoned approach to legislation.

M. Clear, "Falling Into The Gap: The European Union's Data Protection Act and its

Impact On U.S. Law and Commerce" (2000) 18 J. Marshall J. Computer & Info. L. 981.

The European Directive will have dire consequences on the American market if American industry does not adapt itself to the new requirements of the European market.

A. M. Hedgecoe, "Reconstructing Geneticization: a Research Manifesto" (1999) 7 Health L. J. 5-18 .

Geneticization can be viewed in a positive light if it viewed less politically and more commercially. Regulation is needed before geneticization will be accepted.

B. M. Knoppers, "Human Genetics: Parental, Professional and Political Responsibility" (1993) 1 Health L. J. 13-23.

The human genome project represents a massive responsibility that doctors, parents, and politicians need to be prepared to meet. The knowledge produced by the project has the potential to create a culture of genetic determinism and discrimination.

D. Keays, "Patenting DNA and Amino Acid Sequences - An Australian Perspective" (1999) 7 Health L. J. 69-90.

The debate and legislation over patenting of DNA in Australia is compared to the perspectives of Canada, the E.U., and the U.S. The issues surrounding both sides of the debate are reviewed.

B. M. Knoppers, "Reflections: The Challenge of Biotechnology and Public Policy" (2000) 45 McGill L.J. 559-566.

Practical and ethical issues will need to be addressed in the face of scientific advances, and an appropriate policy framework developed. Greater transparency in scientific data and public participation are required in Canadian policy development.

T. Lemmens, "Selective Justice, Genetic Discrimination, and Insurance: Should We Single Out Genes in Our Laws?" (2000) 45 McGill L.J. 347-412.

The completion of the human genome project creates the potential for genetic discrimination, against which Parliament should enact legislation, especially with respect to the existing insurance schemes.

While DNA patenting is still a much debated issue, DNA should be subject to patentability and the relevant authorities should declare such measures to be in force. The implications of DNA patenting are tremendous in light of the human genome project.

To: "Fraser, Claire M." <cmfraser@tigr.org>, "'ksbanerj@uwo.ca'" <ksbanerj@uwo.ca>

Dr. Fraser asked me to respond to your email. I will attempt to answer the multitude of questions and issues that you have posed. In framing these answers, I am representing the perspective of bioinformatics at The Institute for Genomic Research (TIGR), a not-for-profit research institute

that receives greater than 90% of its funding from various granting agenices of the U.S. federal government (i.e. NSF, DOE, NIH, USDA, US Army).

Given the fact that the majority of TIGR's funding comes from the US federal government, there are many guidelines and regulations that are attached to such funding. These guidelines and regulation usually dictate when TIGR is required to release data developed under the funded project to the public (i.e. nightly, monthly, within 60 days, at 3X coverage, etc.). All inventions and discoveries under federally funded projects are subject to the Bayh-Dole Act (P.L. 96-517 and 98-620); 37 CFR Part 401; and 35 USC 200-212. In addition, each granting agency has the authority to include additional terms and conditions in an award that may influence how data, research tools, and copyrighted materials are handled. For instance, in December of 1999, the NIH developed new guidelines and principles on how research tools developed under NIH awards are to be disseminated to the scientific community (you can obtain a copy from NIH's website).

This sets the framework for how TIGR handles the dissemination of genomic sequences data, annotation data, research tools, patentable inventions and copyrighted materials developed under such federally sponsored projects. Overall, TIGR systematically releases genomic sequence data via the TIGR website and GenBank. Some of this data is automatically downloaded each night while other data is released every 30 to 60 days depending on the nature of the project (Claire please correct me if I have misstated anything here). Due to such rapid data release requirements imposed by the federal agencies, TIGR generally does not file patent applications on this sequence data before it is released to the public. I must also note that TIGR's interest is in providing this information to the scientific community as rapidly as possible, therefore, filing patent applications on this sequence data is not a priority for TIGR, nor is there sufficient time to file patent applications. Patent applications are extremely expensive and it takes at least 18 - 36 months for a patent to issue in the field of genomics.

The data available on both TIGR's website and GenBank is searchable, via BLAST search, by both academic and commercial researchers. The BLAST searches only allow a researcher to search one genes sequence at a time and are very time consuming. TIGR has developed licensing terms and conditions that allow academic, not-for-profit, and government scientist/researchers (hereinafter referred to as academic researchers) to obtain access to the flat files for the unfinshed genomic sequence data for genomes currently in progress at TIGR. At the same time, commercial organizations can also non-exclusively license this data for a nominal fee (i.e. $1,000 for a single genome, or $5,000 for access to all of the genomes available under

the license). The terms of both the academic and commercial licenses restrict the licensees from redistributing, transfering, publishing and/or releasing this data; they are site specific licenses; and restricts the licenses from publishing a whole genome paper. These licenses are a mechanism for providing the scientific community with early access to this raw sequence data.

Several members of the bioinformatics department at TIGR have been responsible for authoring new software programs at TIGR. Again, most of these software programs have been developed under federal funds. By virtue of employment at TIGR, the rights in these software programs are assigned to TIGR and each software program is copyrighted (i.e. it contains a copyright

notice) to protect TIGR's rights. Many of these software programs are licenseable from TIGR. Academic researchers can obtain a royalty-free, nonexclusive, nontransferable license to any software program available for licensing from TIGR upon completing a Software License Agreement, Academic Use. These forms are downloadable from TIGR's web site. Commercial

organization can also obtain a nonexclusive, nontransferable, site specific license to these same software programs for nominal license fees. Overall the terms and conditions of both license agreements are the same. TIGR does not provide support and/or maintenance for any of the software programs authored and licensed by TIGR.

Some of the concerns TIGR has faced related to access to the data publicly available on the internet as well as the licensed unfinished genomic sequence data have included 1) researchers trying to publish a whole genome analysis paper before the scientists at TIGR have published their whole genome analysis paper; 2) the unauthorized sharing/redistribution of such data; and 3) proper acknowledgment is not given to TIGR and the respective scientists when other scientist use data obtained from TIGR in their publications. With regard to software, there are concerns that there may be some academic as well as commercial users that are infringing upon TIGR's

copyright in the respective software programs. With regard to commercial organizations, this results in unrealized licensing revenue for TIGR.

Overall, there will never be a perfect system for the dissemination and sharing of scientific data. As a small not-for-profit institute, TIGR, has implemented various procedures on sharing data and information developed at TIGR and such procedures are continually evaluated to ensure that we are making every attempt to fairly and equitably provide such data, research tools, patentable inventions and copyrighted materials to both academic and commercial organizations.

I hope this information is helpful in your assignment. Please excuse any typos, my computer at home does not have spell check for email.

I'm hoping you can help put me in touch with someone who can answer my questions. I'm working for a law professor on legal issues in the world of biotechnology. I have been asked to prepare a brief synopsis of the field of bioinformatics from a legal perspective. I was hoping you might help me with some clarifications.

With respect to the science of the work, I'd like to know what type of legal hurdles (patents, licences, etc.) you face that hinder your work, and likewise, what legal barriers protect your work. How is the information accessed by the commercial world? What legal structures

exist to protect both the scientific and the commercial party? Are there any concerns with regard to the existing legal structures (e.g. regarding access, protection, dissemination of information)? How do these structures detract from the science of the work? Are there any suggestions for improvement to the existing scheme?

I'd appreciate as well the addresses of internet sites that deal with these issues, other research groups, and especially commercial bodies that are involved with bioinformatics for profit. I'd also appreciate any comments you have on etical concerns regarding the legal framework.

[1] For general information on bioinformatics, see National Genomics Research Institute, online: About Bioinformatics <http://www.ncgr.org/bioinformatics/> (date accessed: 4 June 2001).

[2] In genetic studies, "model organisms" have typically been selected to represent wider groups of organisms. For example, Drosophila melanogaster (fruit fly) was an early representative for animals, Escherichia coli represents bacteria, Arabidopsis thaliana represents higher plants, Caenorhabditis elegans represents invertebrates, and the mouse represents mammals.

[3] The past decade saw a massive international push to sequence, or "map", the human genome, which culminated in the publication of a draft in mid-2000. For general history on the human genome project, see online: History of the Human Genome Project <http://www.ornl.gov/hgmis/project/hgp.html> (date accessed: 4 June 2001).

[4] For an overview of the implications of bioinformatics to drug companies and their recent techniques and successes, see K. Howard, “The Bioinformatics Gold Rush” (2000) 7 Scientific American, online: Scientific American <http://www.sciam.com/2000/0700issue/0700howard.html> (date accessed: 4 June 2001).

[5] For an overview of the techniques and how they are used at this stage, see online: Bioinformatics Supercomputing Centre Training <http://www.bioinformatics-canada.org/training.htm> (date accessed: 4 June 2001).

[6] Open Text (online: < http://www.opentext.com/base4/>) is a major commercial producer of software products. The Weizmann Institute (online: < http://www.weizmann.ac.il/>) is a leader in both software development and eduction in its use.

[7] See T. Lemmens, "Selective Justice, Genetic Discrimination, and Insurance: Should We Single Out Genes in Our Laws?" (2000) 45 McGill L.J. 347-412; T. K. Baumann, "Proxy Consent And A National DNA Databank: An Unethical And Discriminatory Combination" 86 Iowa L. Rev. 667; D. Banisar & S. Davies, "Global Trends In Privacy Protection: An International Survey Of Privacy, Data Protection, and Surveillance Laws and Developments" (1999) 18 J. Marshall J. Computer & Info. L. 1.

[8] Celera Genomics (online: <http://www.celera.com>) has effectively competed with the international effort to map the human genome, relying both on the publicly available results of the effort and its own independent research. For a review of Celera's history and the controversy surrounding its competitive stance, see R. Lewis & B. A. Palevitz, "Sequencing Stakes: Celera Genomics Carves Its Niche" (1999) The Scientist 13 (15), online: The Scientist < http://www.the-scientist.com/yr1999/july/palevitz_p1_990719.html> (date accessed: 4 June 2001).

[9] The National Center for Genomics Research (online: <http://www.ncgr.org/>) and The Institute for Genomics Research (online: <http://www.tigr.org/>) represent two of the larger of these institutions.

[10] In the United States, where the majority of this research is conducted, the Bayh-Dole Act (P.L. 96-517 and 98-620), 37 CFR Part 401, and 35 USC 200-212 govern publicly funded projects. Letter from M. J. Brown to K. Banerjee (1 May 2001), reproduced in full at the end of this memo.

[11] Online: The Genome Database <http://gdbwww.gdb.org/>.

[12] Supra note 10. Most scientific journals will not publish articles whose data and results are not made publicly available. This practice has extended to genetic research, where articles concerning genome mapping would be refused for publication unless the sequences were published on the internet. The Genome Database, supra note 11, is largely a result of this practice.

[13] The controversy surrounding private research into genetic information came to a head when Celera Genomics mapped the mouse genome but withheld the information from publication, preferring to treat it as a commercial product. See J. Gillis, "Celera Has Mouse Map Monopoly" The Washington Post (27 April 2001) E1, online: The Washington Post < http://www.washingtonpost.com/wp-dyn/articles/A9192-2001Apr26.html> (date accessed: 4 June 2001).

[14] See S. Eisenberg, "Re-Examining The Role Of Patents In Appropriating The Value Of DNA Sequences" (2000) 49 Emory L.J. 783; C. J. Miller, "Patent Law and Human Genomics" (1998) 26 Cap. U. L. Rev. 893; D. Keays, "Patenting DNA and Amino Acid Sequences - An Australian Perspective" (1999) 7 Health L. J. 69-90.

[15] See J. H. Reichman & Paul F. Uhlir, "Database Protection at the Crossroads: Recent Developments and Their

Impact On Science and Technology" (1999) 14 Berkeley Tech. L.J. 793.