BINF705 FALL15 SOLKA -Research Ethics BINF705 Research Ethics Fall 2010 Ethics at the Interface of...

BINF705 FALL15 SOLKA -Research Ethics

BINF705 Research Ethics Fall 2010

Ethics at the Interface of Computer Science and Biology

Jeff Solka Ph.D.


Acknowledgements Part I

Critical Issues in Bioinformatics and Computing Someswa Kesh, PhD, Professor in the Computer

Information Systems and Wullianallur Raghupathi, PhD, Associate professor of Information Systems

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2047326/

Part II Continued discussions on bioinformatics ethics http://www.dartmouth.edu/~cbbc/courses/bio68/notes/

Ethics_RogerYoung.html

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2047326/


Dilbert of the Week

Critical Issues in Bioinformatics and Computing


What is Bioinformatics? "Bioinformatics" is defined by the National Institutes of

Health as the “research, development, or application of computational tools and approaches for expanding the use of biological, medical, behavioral or health data, including those to acquire, store, organize, archive, analyze, or visualize such data.”1


http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2047326/#B1

The Integrative Nature of Bioinformatics Its interdisciplinary and integrative approach draws

from fields such as mathematics, physics, computer science and engineering, biology, and behavioral science. The generally accepted subdisciplines include (1) development of new algorithms and statistics with which to assess relationships among members of large data sets; (2) analyses and interpretation of various types of sequences, domains, and structures; and (3) development and implementation of tools that enable efficient access and management of different types of information.3–5


http://www.ncbi.nlm.nih.gov/pubmed/11183771

http://www.ncbi.nlm.nih.gov/pubmed/11408660

The Triad Model Computing professionals, including developers,

programmers, consultants, and vendors, should be concerned with building and testing robust applications and performance issues such as correctness of data, reliability, and real-time processing, and integration and management of data deployed to serve multiple purposes simultaneously.

Users, including molecular biologists and other scientists in the life sciences, are concerned with data input and user interfaces, analysis and analytical tools, and interpretation in a shared, global environment. Not only are new genetic tests and procedures required, but clinical information must also be integrated and analyzed continuously to facilitate identification of adverse reactions. Furthermore, the ability to immediately detect promising new drug applications as well as bring new products rapidly to the market are significant challenges. At this time, users are constrained by incomplete, piecemeal tools with poor usability. A primary objective is to work with the other groups in developing sophisticated applications and analytical tools that facilitate fast querying and data mining. The sharing of resources without reinventing the wheel is another challenge.

The public at large is concerned with implications of potential medical applications, ethics, privacy, potential misuse of data, and public and social policies. Almost every citizen is involved in these issues, including social workers, legal and medical professionals, lawmakers, patients, and other participants, including pharmaceutical companies and healthcare providers.


The Triad Model The intersecting area in Figure 1 depicts

overlapping roles and responsibilities of participants in the application of the triad model. For example, the public should decide what can and should be ethical and legal. This will directly place limits on the type of research the user may perform. The user, on the other hand, can and should join this public debate. Once it has been decided what can be done, the relationship between the user and the computing professional comes into play to determine how computing technology can assist the user. This is not a static relationship, but rather a dynamic equilibrium, where the participants in the model will have to decide on the point of equilibrium at a particular point in time and in a specific social context.


http://library.ahima.org/xpedio/groups/public/documents/journal/bok1_024900.pdf

Iceland A law passed in the late 1990s enabled creation of a

medical database that includes medical and family history records as well as genetic information on all Icelanders.

The database was contracted out to a third-party biotechnology firm.

Questions have been raised regarding violations of medical and personal privacy; medical stereotyping of individuals, families, or the entire population; potential discrimination based on medical or genetic data; and a monopoly on medical research and drugs by large companies.


“For Sale: Iceland’s Genetic History,” Oksana Hlodan Who has the right to access and use our personal genetic data? Who controls the data? If medical records are used as a community resource, should they

not be available to all research facilities within the community? Will the medication for a disease discovered through population

genetics studies be available to the participants? Can anybody own pieces of our genome through patents,

copyrights, and so on? Should genetic testing be done, and how scientifically reliable is it? How will other citizens perceive an individual whose genetic tests

reveal a potential disease? Will the data lead to discrimination?


Generalizations of the Triad Model The triad model described here can be generalized for the larger

field of health information management (HIM), which encompasses all aspects of the healthcare industry, including the flow of information therein.

The participants would include patients, healthcare providers (including physicians, nurses, health maintenance organizations [HMOs], insurance companies, hospitals, pharmacies, and medical testing agencies), and federal programs such as Medicare.

The gathering, storage, processing, and dissemination of the disparate and complex medical information generated by the overlapping interaction between these entities will result in the need to address privacy and security issues.

The dynamics of the interaction and the resultant outcomes can be studied using the triad model.


Challenges of Data Formats Regarding data standards, the emergence of the

macromolecular crystallographic information file (mmCIF) and extensible markup language (XML) provides standards that can produce a common format for data.

The essence of XML is this: the problem it solves is not hard, and it does not solve the problem well.

— Phil Wadler, POPL 2003

It is critical that the bioinformatics community either decide on or gravitate toward one common format that will make data sharing vastly easier.


Data Sources Federated database or data warehouse

The three primary sequence databases: GenBank (NCBJ), Nucleotide Sequence Database (EMBC), and the DNA Databank of Japan (DDBJ).

These are repositories for raw sequence data, but each entry is extensively annotated and has a features table to highlight the important prospects of each sequence. The three databases exchange data on a daily basis.8

http://en.wikipedia.org/wiki/List_of_biological_databases


Advanced Methodologies Data mining for patterns is essential for newer discoveries. Pattern recognition algorithms and neural networks have been applied to

bioinformatics research. Neural networks can also be applied to classification as well as decision

problems.11

Other artificial intelligence-based algorithms, like case-based reasoning (CBR), can be useful in this regard.

“Deep computing” for bioinformatics research, implies the use of powerful machines executing sophisticated software based on innovative algorithms to solve complex problems like mapping, modeling, and visualization.

From a hardware perspective, both a supercomputing approach and a distributed computing approach have been used in bioinformatics. Grid computing allows geographically distributed organizations to share applications data and computing resources.12 While the distributed approach is less expensive, it raises further issues endemic to distributed processing and data distribution, particularly those over Internet services.


Data Access GeneX, an example of a system that helps with the storage,

organized retrieval, and analysis of gene expression data. http://www.ncbi.nlm.nih.gov/pubmed/15088382

Among the most important software tools for the understanding of DNA and protein sequences are sequence similarity and alignment tools such as Basic Local Alignment Search Tool (BLAST) and a sequence alignment algorithm using a flat file format known as FASTA. Figure 2 is a sample screen capture of the BLAST interface. One can visualize the complexity of the back-end databases and the front-end query tools with which BLAST deals. These tools allow one to compose an unknown sequence with a database of sequences from other organisms that are better understood. These programs report the hit in the database, along with the estimated statistical significance of the hit.13


Interfaces To facilitate access, several tools have been

developed or are works in progress. These tools include GeneX, an example of a system that helps with the storage, organized retrieval, and analysis of gene expression data. Among the most important software tools for the understanding of DNA and protein sequences are sequence similarity and alignment tools such as Basic Local Alignment Search Tool (BLAST) and a sequence alignment algorithm using a flat file format known as FASTA. Figure 2 is a sample screen capture of the BLAST interface. One can visualize the complexity of the back-end databases and the front-end query tools with which BLAST deals. These tools allow one to compose an unknown sequence with a database of sequences from other organisms that are better understood. These programs report the hit in the database, along with the estimated statistical significance of the hit.13


Operational Requirements Availability—continuous access to the distributed data

warehouse and Web sites Security—appropriate controls for access and information

assurance Data protection—loss of data is decidedly unacceptable,

and backup is critical Data mobility—data need to be available to the right user,

at the right time, in the right place Data purpose—the same data may have multiple purposes

and views Data sharing—access to all information by all participants Real-time availability—data must be available at all times

in a global setting15


Public/Ethical Concerns Bioethics—The moral and ethical implications in the application of bioinformatics to

genetics. For example, is the manipulation of human cells via genetic engineering contrary to the laws of nature and religion? Cloning is yet another issue.

Intellectual property—The ownership of the human genome is probably the most critical issue. Researchers at universities where a great deal of bioinformatics research is done should clarify intellectual property issues with the university. Ownership of the successful experiments performed “in silico” (via the computer chip) is an unresolved question.

Responsibility—Who is responsible for the results? When errors cause injury or damage, who will be responsible?

Access—Who should have access to the data and for what use? Should law enforcement, insurance companies, HMOs, and employers have access?

Privacy—How will privacy be protected? Who controls the information? How will conformance to laws like HIPAA be enforced?

Standards—In terms of gene therapy, what is normal and what is a disability or disorder? Technology access—How will the digital divide between those who do and do not have

access to expensive technologies be reconciled? Outsourcing—How will outsourcing affect the field? Given the sensitive nature of

research in bioinformatics, what additional legal and intellectual property rights issues will develop?


Continued Discussions on Bioinformatics Ethics


What will / has Bioinformatics Enabled Us to do that we couldn’t do Before?

Bioinformatics can predict the future But are these predictions/classifications accurate? Do we know enough about the mechanisms of gene expression

to warrant changes in behavior / action based on microarray data?

If you had a test and result was +6 for enhanced cholesterol synthesizing protein, what would you do?

“The only way to keep your health is to eat what you don't

want, drink what you don't like, and do what you'd rather not.” “Be careful about reading health books. You may die of a

misprint. ” Mark Twain


Medical Concerns Medical records – should bioinformatics data be

included in these? If Included in medical records? Access by whom? Person, md, family, businesses? Justify each.

Bioinformatics and (genetic?) medicine What traits can be identified by bioinformatics?

Introversion versus extroversion.


Testing Presymtomatic testing for genetic diseases –as an adult would you

want to know? Why? Would knowledge of your genotype promote your self-understanding?

Would you then be able to realize your potential, or modify your behavior to adapt to known weaknesses?

If you knew that eg Hunt / BRCA would kill you at age 45, how would that affect your outlook on life?

Justified euthanasia – plan ahead? Huntington’s, parkinson’s, alzheimer’s etc. Incapacitated termination of life. Who should decide, you, your family, the government? Eg -

Florida coma case.


Cures Bioinformatics allows for detection of genetic ‘diseases’ Once you have

detected a ‘disease’, should you cure it? Discuss the ethical issues of curing such ‘diseases’ (ie sickle cell) that are advantageous in other environments

At what cost to society, to px? Benefits? Age of prognosis, severity of disease. Examples of obvious cure, and then less so (CF, MS, Alz, Hunt, blind,

deafness, intelligence, reproduction, colorblindness (job limiting), body build, height, hair baldness, eye color …..)

Class to define, & justify, treatable diseases. What is the impact of an individual being able to know their ‘at risk’ genes on

demand ($500 and 48 hrs?) Psychological Employment Health care Social / relations

https://www.23andme.com http://dna.ancestry.com


Medical Training Should new technologies – such as bioinformatics – be regulated or adopted in

the medical world? Should MDs be familiar with bioinformatics? How will MDs cope / deal with this extra information. How have they handled

past increases of information? (PCR, lab tests, new drugs…) Is this just too much information for mds.

How much bioinformatics should be taught to medical students? Who here is premed? How are you going to use bioinformatics info?

Is the current system of health care in the US prepared for such information Is the concept of health ‘insurance’ outmoded here where individuals are

denied / accepted by insurance companies based on their predisposed risk to genetically based medical problems?

Who will be ‘at risk’? (poss instant role of ins company & family) NB – health insurance is a factor in US mainly.

http://genemednareport.com


Public Self Perception and Acceptance Case – gm food – pros & cons. What has been the reaction of

public to gm food? Too complex for public? Scared of the unknown? What is a ‘normal’ human genotype? Does different = wrong How different is wrong enough?

Who want to admit to their mutations? Color blind, balding, myopic, …….

What is the public / politician perception of ‘normal’ How does the public/a group react to someone/another group who is ‘different’? – scared, hostile.

Will ‘difference’ become overused and lose it’s stigma, as everyone is classed as ‘different’ to everyone else? How will public perception of ‘difference’ alter over next decades with constant use / referral to Human Genome Project & bioinformatics data where no-one is the ‘same’.


Stereotyping What are the ethical, legal and social implications of assigning such

genetic/sequence data to individuals, groups, cultures and societies around the world.

Will groups be evaluated based on their genetic profile? will populations / cultures / individuals be identified by their genotype, rather than their phenotype (as it is now)? will this increase / decrease racism/sexism etc?

What is the genetic difference between different human cultures? Pick any two & hypothesize on differences – phenotypically, biochemically.

Once an individual’s genotype is easily available, will there be genetic discrimination? Is there such discrimination now?

Are their really different groups of people / individuals, or are we all really just part of a continuum? Example of height, iq? Skin color, hair color, eye color.

If we consider 2 individuals of same culture/population – what pheno / biochem diffs are there?

Some examples might include allergies, lactose, other foods, diabetes, color blindness, flat feet, !!!


Stereotyping What about epigenetic effects?

What about horizontal gene transfer


Implications of Biometrics Transponder implant with credit card info. Pets with id chips. Is this the way of future medicine? Is the prognosis with chip containing medical data –

safer, no mixups, scanned, monitored. Current pacemakers – checked by computer over

phone. What about hacking insulin pumps and pacemakers? http://www.reuters.com/article/2013/07/26/us-

hacker-death-idUSBRE96P0K120130726 Will people be easily scanned/tagged this way? Individuality, ‘freedom’? ACLU?


Implications to the Workplace Are you defined by your genetic sequence? Will employers require a specific genotype / gene in

employees? Can they request this like a drug test? What is a ‘minority’? How is it defined? How is it likely to

change in a society where genotypes are easily determined and where everyone is a ‘minority’ with regard to a certain gene?

Should we base ‘minority’ status upon the degree of deviation from a genetic ‘norm’ – i.e., the more mutations you have, and are therefore more different to the ‘norm’, the greater the claim you have to minority status and special / advantageous treatment.

How many mutations, and to what extent? Should we measure genotype or phenotype?


What is a Disability and What is Normal What is a ‘disability’ when the same genetic change may confer alternative

advantages? How is it defined? How is it likely to change in a society where genotypes are

easily determined? What happens if, for example, there is a genetic test for ADD? Will some people

with the phenotype be diagnosed as without the genotype (as this is very much a subjective ‘disease’) and therefore not be eligible for health coverage on this, or will some people be diagnosed with the genotype, but not the phenotype and therefore be stigmatized with a ‘disease’ that they do not show any symptoms for?

What will be the actions of the pharmaceutical companies that make ‘ridalin’? What is in their best interest - how will they act politically?

What are the implications of separating genotype and phenotype on discrimination?

Currently, we discriminate based on looks, right? Do we discriminate based on data? (wealth, power, connections, titles?

Phds? Now genotype?) Example - BRCA1 predisposition, Huntington’s gene, Alzheimer’s etc, intelligence genes.


Abilities Find the genes for intelligence. Would you be pre-selected for schools

like Dartmouth? Would you want YOUR children to have extra intelligence genes / be positively/negatively selected based on their genotype

Docherty, S. J., Davis, O. S. P., Kovas, Y., Meaburn, E. L., Dale, P. S., Petrill, S. A., ... & Plomin, R. (2010). A genome‐wide association study identifies multiple loci associated with mathematics ability and disability. Genes, Brain and Behavior, 9(2), 234-247.

Sports, enhancement ACTN3 Genotype Is Associated with Human Elite Athletic Performance Nan Yang,1 Daniel G. MacArthur,1,2 Jason P. Gulbin,3 Allan G. Hahn,3 Alan H.

Beggs,5 Simon Easteal,4and Kathryn North1,2

Looks / beauty – subjective across groups / cultures.


The Esoteric Aspects of This Science If everyone of a particular species is slightly different genetically, how

much genetic difference is required to be a different strain / race / sub species / species / etc.

What experiments / analyses would you perform to answer these questions?

Perform phylogenetic tree analyses on wild type and mutant genes in humans (ie cftr, brca1, apoe3/4 etc). note degree of change and compare amount to that seen between a single gene between two different species.

How do you distinguish between several/many mutations that have no effect, and one mutation that has a large phenotypic effect. Which one is more ‘different’?


Bioinformatics and Pharmaceuticals Can you tailor drugs to work with a specific genotype?

Should the genotype of patient be known before taking a certain drug?

Does this enhance the effectiveness of the drug, or invade the privacy of the patient?


Bioinformatics and Gene Therapy Will the ability to quickly, accurately and cheaply

determine an individual genotype promote gene therapy as a medical approach?

What other technological barriers need to be overcome before this information is medically helpful? Ie – what are the current barriers to gene therapy?

Future developments Can we alter / enhance / prevent certain

developmental stages? Replacement organs? Ears, skin, kidneys …


Bioinformatics and Evolution - I Will bioinformatics help fill in the ‘gaps’ in our evolutionary knowledge?

Can we predict what gene & proteins might look like/function in species that we have no evidence for? The missing link?

Evolution studies are currently based on features or traits - should it be revised to be a genetic reflection of our ancestry? Ie bergey’s manual.

Linking hard data with population genetics P2+2pq+q2 predicts, based on statistics. bioinformatics will allow

instant screening & real numbers. If the genetic data from a population can readily be analyzed (the real

data, not just probabilities as before), can / should the genetic future of the population be predicted?

Should that future be manipulated for the ‘good’ of that society? (Obesity, cancer, diseases due to inbreeding (Ashkenazi Jews & Tay Sachs is an example of where this works in a pre-bioinformatics era).

What about introversion? What about autism or Ausbergers like symptoms?


Bioinformatics and Evolution - II Darwinian theory, natural selection, survival of the

fittest, does this need to be modified? What is the ‘fittest’? What does this mean from a sequence data point of view? Least mutation? Most mutation?

With genetic information easily available, will this alter the evolution of the human species?

Will partners be chosen based on genotype rather than phenotype?

How are partners chosen now?


Case Studies (Hospital) Will you commence prescreening of fetuses for use in

prenatal (pregnancy) counseling?

What do you need to know before doing this? What are the needs/gains for the px and the costs/gains for the hospital?

Gen counselor, hospital lawyer, GM of hospital, parents,

unfaithful wife!, husband, parents of husband with history of heart ds


Case Studies (United Airlines) Screen pilot applicants for neurological issues /

addiction genes / depression / heart problems etc etc = anything.

Why would you want to do this?

Safety of of passengers is at stake. Can UA require this? Safety = increased costs. But better PR (our pilots are safer than yours!)

Pilots, UA gm, lawyers for each party, passengers,

geneticists, mds.


Case Studies (School Bus Drivers) School bus drivers Applicant screened for drugs etc – also genetic

screening – show predisposition for stroke, poss danger to children if driving bus? No symptoms

Applicant, school district manager, school lawyer, union leader, ACLU representative, spouse of applicant (50 yrs married, no problems, healthy), children (need more drivers, w/o = waiting increased for children, reduced bus routes)


BINF705 FALL15 SOLKA -Research Ethics BINF705 Research Ethics Fall 2010 Ethics at the Interface of...

Documents

Transcript of BINF705 FALL15 SOLKA -Research Ethics BINF705 Research Ethics Fall 2010 Ethics at the Interface of...