Nanguiano Individual Assessement
Statement of Work
On my project, I completed a successful export of V. cholerae which solved the problems I had faced in the initial export I had attempted. To perform the export, I modified an existing GO OBO-XML DTD schema to work with a new OBO-XML file that used a schema that no longer worked with GenMAPP Builder. I ran xsd2db on the manually edited DTD file to obtain files that could be used in GenMAPP Builder. I ran GODB Post Processor on the specified HBM and SQL files to allow them to be properly added to GenMAPP Builder. I replaced the old files in GenMAPP Builder with the new files generated by xsd2db and GODBPostProcessor, and overwrote GMBuilder.sql with the contents of the schema.sql file that was edited by GODB Post Processor so that it could build the proper table initially. I wrote documentation on how to perform updates to GMBuilder in the event of an OBO-XML schema change in the future. Following all of this, I performed an export of V. cholerae, which completed successfully. Then I performed quality assurance on the export, the details of which can be found in the
Report
. Following this, I converted the XMLPipeDB website to Jekyll, and moved it to a gh-pages branch on github, giving it a new github domain. The primary website was redirected to the github site link. Following this, I ran SchemaSpy on the database, the results of which can be found here for uniprotdb, here for godb, and here for gmbuilder. I also wrote up documentation on how to use SchemaSpy for future reference.
Assessment of Project
Overall, I would say the project was a success. The initial goal for the semester was completed, and the following goals of converting the XMLPipeDB website to Jekyll and moving it to github were also complete. The most successful aspects of the project were the parts that combined my knowledge with that of Dr. Dionisio and Dr. Dahlquist. Many of the tasks I completed this semester worked with things I'd never worked with before, like ant, hibernate, and xml. Attempting to complete work on my own was often fruitless, as not knowing what I was doing often made it difficult to search for precisely what I needed to do. Working with the professors to learn what they learn instead of trying to do it all myself allowed things to progress much more quickly. If I were to do it again, I would come to them for aid much sooner than I did so as to avoid days of struggle attempting to figure out errors that were very easily recognizable by others.
I am very pleased with the results of my work on the V. cholera export (and corresponding edits to GMBuilder), the XMLPipeDB site, and the SchemaSpy documentation. The export and quality assurance performed seemed to indicate that the edits to GMBuilder will allow for future GO OBO-XML files to run smoothly without error, provided another DTD schema change does not occur. However, should one occur in the future, the documentation written should make the process much smoother for future developers. Due to the use of Github, I would say the project was very organized. Tasks and accomplishments were clearly marked, and questions and concerns were generally kept in their relevant threads. The entire process from beginning to end is documented fully on github's issues, in issues both opened and closed as of the time of this writing. Additionally, all code written and changed can be found in the github repository for XMLPipeDB. It is my belief that Github allowed for this project to run very smoothly and for all correspondences to be recorded without confusion.
