The Mouse Diversity Panel (MDP) uses data downloaded from the Jax Phenome DB in June 2006. The data are stored in the Phenotype table of our MySQL database, and their publication records are held in the (unsurprisingly) Publication MySQL table. There is additional data that may be added at some point in the future, but as of now the only data stored from the downloads are as follows:

Study Name, Units, Abbreviation (in Phenotype table) and Authors, MPD_ID & Short Study Description, Year (in Publication DB - also many null values)

There are many known bugs with our implementation of the MDP at the moment, one major problem being the lack of centimorgan information, which leads to an inability to create our genetic maps.

A list of standing bugs is as follows, continuing from the Roundup issue. I have removed all fixed bugs. Comments are in plain text between each issue.

1. We need to link from each trait in GN directly to the correct record at JAX. For example, our Record ID 36107 "age at start of KCL test (male)" needs to link directly to

http://phenome.jax.org/pub-cgi/phenome/mpdcgi?rtn=assays/onesumm&measnum=6107

I think Jintao said he just added 30000 to all of the JAX ID numbers to generate our Record IDs JAX MDP 6107 = GN Record ID 30000 + 6107 = 36107

These links should be made available on the Trait Data and Analysis form.

The records that had a JAX MDP ID of > 10,000 before being imported had 10,000 added to their number, the ones < 10,000 had 30,000 added. I believe this was done so as to avoid duplicate ID entries in the MySQL table into which they were imported (I don't know which table yet, 08-02-06). I'm confused as to why a simple, constant number was not applied to all of them. The links can be fixed using a few simple lines of code, but it would be very ugly. However, unless we can easily port the numbers over to a consistent change, that's probably what will be done.

6. Did we manage to integrate the "N mice" data for all of these traits? We do have a place for these data in our MySQL database. JAX displays these data and we should too at some point.

This data is in an Excel spreadsheet on Manly G5. It is not in the MySQL database however, nor is there a column for it. It could be added without major difficulty (but probably a fair time investment).

7. JAX also has something called "Project Title", "Category", and "Units". do we have these data. If not, we should get these data one way or another. Units is already in our MySQL DB, but not being displayed for some unknown reason. Project Title is being displayed in a shortened form, but the longhand is also in our database. Category is not in either the MySQL db nor Excel spreadsheet. This data is not easily available for download from Jax.

8. Evan, we probably have not yet gotten hold of the "Protocols" for all of the JAX data. This would be very valuable text data to download, since it describes what these data actually mean (the metadata). This is the link to the protocol for the KCl data.

http://phenome.jax.org/pub-cgi/phenome/mpdcgi?rtn=prodocs/61#KCl_age

Please get all of these data. There should only be about 100 of these protocols, maybe fewer.

Again not available for easy download, but if there really are only 100 it could be done by hand (as the Publication table was generated) probably about as quickly as by script or calling Jax.

9. Please test drive the MDP and find link errors. This morning, none of the Record ID links in the Correlation Table for 36107 actually link. They generate the following error:

Mod_python error: "PythonHandler WebQTL? "

Traceback (most recent call last):

File "/usr/lib/python2.3/site-packages/mod_python/apache.py", line 299, in HandlerDispatch? result = object(req)

File "/node_apps/www/html/webqtl/beta/WebQTL.py", line 148, in handler page = dataPage.ShowDatabasePage(fd)

File "/node_apps/www/html/webqtl/beta/dataPage.py", line 1036, in init DataEditingPage? .__init__(self, fd, thisTrait)

File "/node_apps/www/html/webqtl/beta/dataPage.py", line 122, in init self.dispSubmitButtons(fd, title2Body, thisTrait)

File "/node_apps/www/html/webqtl/beta/dataPage.py", line 509, in dispSubmitButtons newWinText = HT.Paragraph('To analyze a trait, select appropriate options and one or more TypeError? : expected string or Unicode object, Chromosome found

Not sure what the problem is. The SNPs should work fine. Will have to look into further.

10. Interval Mapping: We know thisdoes not work,but the client will not know this. Clicking on this button now generates the follwoing error, butit might be better to figure out a better long term fix.

Traceback (most recent call last): File "/node_apps/www/html/webqtl/beta/webqtlCmdLine.py", line 122, in ? page = IntervalMappingPage? (fd) File "/node_apps/www/html/webqtl/beta/intervalPage.py", line 382, in init gifmap = self.plotIntMapping(fd, intCanvas, startMb = self.startMb, endMb = self.endMb, showLocusForm= showLocusForm) File "/node_apps/www/html/webqtl/beta/intervalPage.py", line 547, in plotIntMapping plotXScale = self.drawGraphBackground(canvas, gifmap, offset=newoffset, zoom= zoom, startMb=startMb, endMb = endMb) File "/node_apps/www/html/webqtl/beta/intervalPage.py", line 1591, in drawGraphBackground canvas.drawRect(startPosX, yTopOffset, startPosX + self.ChrLengthDistList[i]*plotXScale, IndexError? : list index out of range

Can be removed, or a simple warning can be issued saying it will not work. I'd be in favorof simply removing it.

11. Marker Regression works great, but we do not put up any intermediate page that tells the client to hold tight while we compute the answer.

Not fixed, shouldn't be hard to use the same code from one of the other searches.

12. The link to Genotypes in the Marker Regression Report generates the following link/Python error:

Mod_python error: "PythonHandler WebQTL? "

Traceback (most recent call last):

File "/usr/lib/python2.3/site-packages/mod_python/apache.py", line 299, in HandlerDispatch? result = object(req)

File "/node_apps/www/html/webqtl/beta/WebQTL.py", line 148, in handler page = dataPage.ShowDatabasePage(fd)

File "/node_apps/www/html/webqtl/beta/dataPage.py", line 942, in init thisTrait = webqtlTrait.Trait(db=database, name=ProbeSetID, cellid= CellID? , cursor=self.cursor)

File "/node_apps/www/html/webqtl/beta/webqtlTrait.py", line 170, in init self.db = Database(self.db, self.cursor)

File "/node_apps/www/html/webqtl/beta/webqtlTrait.py", line 63, in init self.retrieveName()

File "/node_apps/www/html/webqtl/beta/webqtlTrait.py", line 128, in retrieveName raise KeyError? , `self.name`+' doesn\'t exist.'

KeyError? : "'MDPGeno' doesn't exist."

I think this is a similar problem as 10, but again not sure exactly what's causing it. It might even just be a missing period (should be MDP.geno). Again, something I'll have to look into.

13. The data in AXB820, 19, and 18 have to be collapsed.

This may already be done automatically, I am not sure. I think it is, but I do not know how to check.

-- EvanW - 01 Aug 2006