Projects: GENIQUEST : Randy's initial storyline proposal

RVS 20071119
Dragon phase.
I'll start with just the first segment, the Dragon/Drake phase, of the program. The human/mouse segment can be modeled the same.

First let's just list the Core principles we what to cover in our presentation:
Introduction: Most of the high school projects I have seen over the years are based on the concept of formulation of a hypothesis and then designing an experiment to test that hypothesis. Although this is the "scientific method" this is an oversimplification of how science is done in the real world. We should include the social aspects of science.

(1) Model on how science is really done. Include identification of socially important problems, writing proposals, understanding budgets, referencing literature, peer review, writing reports, giving presentations, etc.

Science is not done in isolation. Grants from industry and government are driven by social need. Science is built on the foundation of prior work and the resources to explore and mine this vast pool or information is critical to any successful project. The next step in the process it a blend of science and salesmanship, Proposal Writing. Here the scientist must show a systematic method for solving a problem and outline a path that minimizes the risk of failure. With the proposal comes the budget. Yes it comes down to money and with limited resources we need to make smart choices on how we spend both our time and our money. Good science is often a combination of good ideas and great money management. Next comes the scientific investigation itself, which seldom follows exactly the full course of the proposal and it is the interesting turns that can be the most fun, solving the problems as they come up. Finally, after the answer is found, you need to present your results to your peers and the public and provide enough proof to show that the problem you tackled is solved.
(2) Incorporate the principles of reproducible research. Record what was done during a session. Include a research notebook for the students and teams to record their ideas and their work.

These are included in the underlying model for the Independent Studies in Computational Biology year long course that is being taught by the Center for Genome Dynamics and I am confident that we can incorporate these principles into a more condensed model for use in high school and undergraduate studies.

I am going to describe a scenario to open discussion. I am not going to take into consideration feasibility of the programming or resources. We can discuss this separately. I will include comments on potential implementation in braces { }.

Model: a problem solving adventure computer game composed of a mixture of text interaction (like the old Zork games), web pages (a library), a diary/research notebook to record all activities (WIKI), and interactive graphical packages (a virtual lab, ).

Phase One: You have been hired as an apprentice Dragon breeder and this is your first day on the job. You meet the old master of the Dragon keep and he assigns you to explore the archive of Dragon lore. You join a team of fellow apprentices and the team is asked to review the breeding records of the dragons of the keep and identify a set of "traits" and discover how they are inherited.
Deliverables: Each team will be assigned a set of traits. They will have to determine if they are recessive, dominant, epistatic, sex linked, etc. These results are handed in for evaluation. Optionally a symposium can be given for poster and/or oral presentations of results. Notebook should also be reviewed and evaluated.

Time: Depending on the number of traits this segment could be one to two class periods.

<em>In implementation, the current dragon program would work if we limit the breeding genealogy to a set history. This would mimic a set of breeding records like those keep by a horse stable. Not all options for breeding will be available from the genealogy (because in history they did not occur) but enough should be present that the students can determine the traits. We would introduce Drakes as a model for Dragons at this stage. Drakes are small and breed fast, but they have 98% the same genes as Dragons and their genomes can be mapped to each other.</em>
Phase Two: There has been a group of emergencies in the community, Dragons are getting sick. The only think that is known is that it appears to be inheritable. There are eight know strains of Dragon (valley, mountain, desert, ice, ocean, river, swamp, cloud) and several critical diseases that only some of the strains catch. The master of the Dragon Keep gives the apprentices access to the archive of all the knowledge stored on the genetics of Dragons and Drakes. After review of the problem and the methods in the literature the apprentices must come up with a proposal and a budget for solving the problem.

Deliverables: The proposal and budget. As an option the proposals can be presented at a symposium for peer review and comment by the rest of the investigators before approval and implementation.

Time: Depending on if a symposium is presented for the class, two to three class periods.

Unknown macro: {In implementation this is where we introduce the concept of breeding strategies, markers (locations on the chromosome that we can test to see if they belong to Strain1 or Strain 2, SNPs (Single Nucleotide Polymorphisms, this is a single base pair change that may be different between two strains), QTL (Quantitative Trait Loci, a location on the genome that is correlated to an observable phenotype, eQTL (expression QTL, what changes in gene expression are observed from microarray data), haplotypes (the specific genetic information that results from the descending through the pedigree of multiple strains [some strains are "closer" than others and share more SNPs], consomics (animals that have one chromosomes from different strain, congenic (animals that have a small section of one of their genes replaced by that section of a different strain), targeted knockouts (a technique that allows you to selectively deactivate a gene), GO (gene ontology, a controlled vocabulary that is used to describe a genes know function), etc. All of these in short articles in the archive (one page or less) with time and cost estimates. These need to be seeded in with other techniques that will not be feasible on the basis of time or money (full sequencing of the Dragon genome...). This archive is just a collection of web pages of techniques and cost estimates and at this time does not exist in the current software.}

Phase Three: * *The apprentices conduct their research by implementing the breeding scheme for the Drakes, collecting phenotype and genotype data on 200 to 300 Drakes. They send this data to Rqtl to get QTL graphs and isolate regions of the genome that are associated with the phenotype. They then go back to the literature to determine that genes in that region and explore their haplotypes and gene ontology to isolate the most likely cause. They can then search for selective consomics and congenics in the literature we provide to see if any overlap with their region and finally use targeted knockouts to confirm their selection. They write a final report.

Deliverables: The report and how well they stayed within their budget. As an option the results can be presented at a symposium. I strongly support this final symposium as it will give the students a change to share the excitement of solving some pretty tough problems. Their diary/notebook is reviewed to see how well they documented their process.

Time: Depending on if a symposium is presented for the class, two to three class periods.

Unknown macro: {In implementation, we will need to include about 30 markers on each of the Drakes chromosomes. In addition we can include additional SNP markers (about 100 per chromosome should do) so that each gene has at least one or two SNPs. The markers are needed for the QTL analysis and the SNPs will be important for haplotype mapping. It would help if we had a genome browser. There are several open source genome browsers that we could adapt for this use (GBrowse in Perl comes to the top of the list. We have some experience with this and have customized Gbrowse for our own web site). They will need resources from the archive of data we give them, guidance from their teachers, and working in a team environment (it will be hard for a single person to search all the areas). From our side, we will need to provide them with a way of viewing haplotyes and congenics on a genome browser so that they can isolate potential genes that cause the problem. Then we need to have a way for them to do a targeted knockout (give a unique gene sequence they want a probe built from that will turn off a gene. Some knockouts can be build so that the gene can be turned on or off with the addition of a drug in the food supply). Somehow we need to send them a report of the results of the knockout experiment they suggest.}

What did we present to the students?
(1) Overview of how science works and its place in society.

(3) An overview of techniques used in genetics to isolate a region of the genome that corresponds to an observable phenotype.

(b) QTL:Quantitative Trait Loci for determining genome locations that correspond to an observable phenotype

(d) haplotype analysis: using genetic information descending through a pedigree of strains to reduce the list of candidate genes.

(e) consomics and congenics and how they can be used to reduce the list of candidate genes.

(f) GO: gene ontology and how genes are annotated and described using a controlled vocabulary.

(g) Targeted knockouts and how they can be used to selectively turn off a genes expression.
(4) How to write scientific proposals and reports. How to present results at a symposium.

(5) How to work in a team and divide labor. How to maintain a budget for time and cost of your work.