Techniques and technologies that have been used for almost a decade on the internet have been creeping into the scientific research world at a slow pace.  An example of a technology I gave was using RSS to keep track the vast amount of information on the Internet, which thinking about the reaction I got was clearly a surprise hit. (To further the point, I’m not the only one who received the same reaction to the same type of crowd.)

A distinct point I made was by using the data our own web site traffic, I found a large percentage of it was coming from search engines not looking for the science we do, but interested in the people that are involved with our work. I’ve since revised this, but I still stand by my original point: who you are and the identity you have online will be just as if not already more than important as the papers you publish, the school you go to, the grades you get and the personal connections you make.

At the end, I wanted to make a point very clear by demonstrating that the ideas I was presenting was nothing novel or outrageous and the data records I was presenting about our own website traffic should not be as surprising at it it was.  Some people such as Anil Dash predicted just how the phenomenon of the internet would work seven years ago.

Here comes the catalyst

During his campaign for the presidency, President Obama orated strongly that utilizing modern technology and Internet techniques was going to be critical in the future progress of America.  So much that he believed this that he hired one of the founders of Facebook to head and organize his campaign online, which has been debated as one of the most important factors in his election.  Since being elected, the executive and bureaucratic arms of the American government have been working hard on the Open Government initiative.  By providing access to information in an attractive, easy to use fashion and data using free, open, non-proprietary technologies to me, personally, is astounding, especially in the rate it has been happening.

What does this mean?

Recall Anil Dash, often-quoted expert of the internet and emerging social technologies.  The rest, I’ll let him explain personally:

Again, this is not an extravagantly new idea.  In fact it mirrors our efforts with FaceBase and CTSA involvement but differs on an important but subtle point: the Expert Labs initiative attempts to find experts (people) with problems (data), whereas other efforts have concentrated on the traditional idea of helping the experts (people) find solutions to their problems (data).

Take the point that I typically bring up when discussing this issue.  A co-investigator and myself have been working off and on for a couple years now utilizing the new PolyPhred utility to re-analyze our mountains of sequence data.  However, we were unsure on how to utilize the data the new version generations and as such had to write a bunch of utilities ourselves.  My first question is ‘are we the only one doing this’?  Then answer is: I have no idea, a Google search brings up the Polyphred site, the original Polyphred paper in published in 1997, some tutorials for using it for Mac OS9 circa 2000, and another paper in 2008 about the new version’s capabilities.  We may very well be the only people doing what we are doing with Polyphred, and in effect we should be able to share how we are doing so, things that typically do not get published but sometimes added as a supplemental material if at all.  This problem was one of the ideas that drove me to thinking about our new web site structure, actually.

It’s also in this point that I hope that something like Expert Labs works, but seeing the pedigree, and the leadership behind it, I have a strong feeling that it will, and something definitely to keep an eye on.  Which, being a modern technology driven initative, they certainly do, with RSS feeds and Twitter updates.

With that I quickly dismantled the old content site and replaced it with Wordpress, hoping that it won’t affect anything else too greatly.  I’ve done quite a bit of work in the past couple years to try and make sure that didn’t happen, but I was insecure about how well I did it.  However I knew that the original content site was insecure, and overall it was quite the insecure situation, which doesn’t help with personal insecurities!

In the end I didn’t have any choice and had to run with it, expecting to shake out any potential problems that could come up.  Other parts of our system appear to be working fine right now as far as I can tell, so overall a minor victory.  The lesson learned here is if there is something you need to do (manage a website) that’s already well defined (through Wordpress, Drupal and others), there’s no need to duplicate work.

With that, hello, and welcome!  Things are barren at the moment as we move through an abrupt and unexpected transition.  It’s going to take Google a little bit to reindex the layout of the new site, so links coming in from Google will be broken until that happens.  One of the best things about Wordpress is that the structure it automatically generates is Google friendly, so as we add content Google will be able to quickly find and index it, and thus the world knows a little more about us.

If you need to change an existing field into a computed field in progeny, please start the computed field as a completely new field and delete the old one rather than changing the field type on an existing field.  When you change the field type of an existing field to a computed field it will cause a database error.  The field won’t work and it will interfere with things like how the individual or sample data sheets are displayed or how LIMS actions/reactions that utilize that field work.  If it happens, we just need to delete the field everywhere that it is referenced, so it isn’t the end of the world, but it is better if we just avoid the problem all together!  Thanks.

Anyone can come to me with additional questions, but the thing I wanted to remind everyone about is performing error checks.  When you select to check for mendelian errors, or discrepancies you should check the corresponding tab for those types of errors at the end of the import to see if you have any errors and to resolve them if possible.  (See screen shot below)

You should always check for:

1) Rejected genotypes:  This will tell you if some of your genotypes were not entered.  (examples of possible reasons include a typo in the sample or marker name or your marker is not in the set you selected to import into)

2) Discrepancies:  If someone else has already typed the marker you are typing or if you re-run a plate or re-process a plate and re-import, there may be differences in what was in progeny and what you are trying to put in.  When you select to check for discrepancies Progeny will keep the original data unless you resolve the discrepancy. I am telling you this because someone recently incorrectly processed plates and imported, then re-processed and re-imported, but was still getting the original (wrong) genotypes out for analysis.  The reason was that the original genotypes were still in the database and the new (discrepant) genotypes were being held, waiting for confirmation of whether or not they should overwrite the existing genotypes. In order to overwrite existing genotypes, you must either a) resolve discrepancies, or b) not select to check for discrepancies (new genotypes will automatically overwrite what is in progeny).

Here is an example of the genotype import page.  Notice the tabs across the bottom with the different types of error checks that you should review. The Progeny user manual also has more in depth information about this.  Feel free to ask me any questions about this.

Jamie