I finished scanning my best microarray to date.  After doing 5 - 6 arrays per week, it finally looks like I'm getting somewhere.  Now I just have to reproduce it.

I was looking up the patent filing for coupling fluorescent dye to amino-allyl groups today, trying to find the exact chemical reaction so I can troubleshoot some of the problems (it's here if you're interested).  I noticed it was filed on July 22, 1997 (my birthday).  I realize that there's a pretty high chance this could happen - 1/365 - but it amuses me when it does.

I've been redoing my microarray preparation all week with genomic DNA - basically all the DNA from the cells randomly broken up by sonication.  Then I label it with fluorescent dyes and let it bind to the spots on the array.   With this technique, almost all of the spots on the array should be yellow since the two dye colors are the same and they should bind to any of the spots.

I had problems the first two times with the spots not being very bright and everything not binding very well.  Part of the protocol I follow involves using a kit from Qiagen to clean up the product.  One of the buffers I was using is a weird yellow color so that it tells you if the pH is off.  Well, we ran out of the kit earlier in the week, so I got a new one from the supply center.  In it is a note that says if you use the kit with microarrays, the yellow dye in the kit can interfere with the fluorescent dyes.  I called them and they confirmed that it has been a problem with other researchers and they overnighted me a bottle of the old, clear buffer without the stupid indicator dye.

I used that in the prep I started on Wednesday and I got much better results.  Not great, but definitely an improvement.  I'm a little annoyed about the situation and that they didn't put it in the product literature instead of on a sheet of paper that is more likely to get thrown away.  But, they did correct the situation quickly and were very helpful.

I did my first microarray hybridizations this week. That means I bound two different fluorescently labeled DNA samples to the slides. When done correctly, it results in something that looks more like the userpic. Unfortunately, it didn't go so well and I had problems. It's not too surprising, given that it's the first time I've done it on my own. It is a little frustrating because I wanted it to work out better. I am going to try again next week to get this working correctly with some input from the expert in my lab. Hopefully, those reactions will work out better.

I started the array printing process this morning.  Basically, I'm putting on 49,152 distinct short DNA sequences that represent the human genome (and a bunch of controls) that I'll later be able to use to see what genes are turned on or turned off in a given sample.   I'm on plate 3 of 128.  Only 39 more hours to go!

I'm working on printing spotted microarrays to use in my research.  We have a robot in the lab to put all the genes on slides.  I'm putting just under 50,000 spots on a normal-sized microscope slide.  It's pretty impressive, although it is a little tedious to get the robot properly calibrated and printing all of the spots.  I also have to keep the humidity at around 40%, so it is completely sealed off from the ventilation system.  I'm spending my days in room that's 29* C and 44% humidity.  Yep, I'm working in a sauna.  Here's the robot we use: