black boxes

We all have black boxes in our lives. Things that work, that we don’t really understand. My car is one. I put gas in it, it takes me places. I don’t really understand how it works. (Maybe this is why two years ago I needed a whole new engine because apparently you are supposed to change the oil?) Ergo, black boxes are fine, until they’re not. And then it would be helpful to know something about them.

As a technician, sequencing was this huge black box for me. I handed off my libraries, they got sequenced, analyzed, and then we back a few files, some graphs, some RPKMs, and that was it. So this is why I decided to to go grad school, right? So I could spend some quality time with SAMtools, figure out what was going on.

It turns out (it always does, doesn’t it?) that I got more than I bargained for. I have a two-week assignment where on Monday, the prof handed us a link to a reference sequence on some online database, two fastq files (raw sequencing reads) and told us to write a perl script to align these reads, give us number of snps, number of indels, etc.

LITERALLY that is all he gave us. Like, perl? I have never used perl. I’m not even sure if I’m supposed to be capitalizing perl. (Ed note: you are.) But hey, I wanted to see what was inside this black box, is there any better way than just being dropped inside of it?

I was thinking about another project for another rotation, which might involve primer design. Primer design is less of a black box, because there are programs where you input the piece of DNA you want to amplify, and the program spits out the ideal primers (taking into account GC content, annealing temperatures and that sort of thing) for that piece of DNA. Next, you take your primers, and do an in-silico PCR, where you go into another program, pick your reference DNA, put your primers in, and it will tell you if your primers map to multiple places on the genome.

So I was walking to my car yesterday, and I was like WAIT. WHAT IF I CAN’T DO THIS. Because if my genomes don’t have a good reference and I don’t really know what’s going on, I can’t do an in silico PCR and what if my primers are horribly designed and map to a thousand places on the genome!

And it hit me. (Not unlike a car, whose engine has run out of oil, this was accompanied by a large clunking sound) I...could write a program that does this. I mean, I could input my own reference. Put in my own primers. Require some specificity, require some distance apart, return the sequence in between, etc. I could probably even email the guy, ask for his code, and put in my reference sequence. Because in programming, that's not plagiarism, that's encouraged! (Sorry I'm totally getting this printed on a shirt at some point.)

Now, this isn’t incredibly groundbreaking. But what’s crazy to me is that I would have never imagined that this was an option before now. Three months ago, I wouldn't have even considered this as an option. To be honest, I would have just winged it, ordered my primers, done the PCRs, hoped that I only got one band and that band was the one I wanted. And that probably would have been fine. But I love that this is an option. I love that I'm learning to think like this. Look! There is one less black box in my life!

**

The other black box in my life is this one. Lena gets on the bus at 7:55 every morning. School starts at 9. What the heck does she do between 8 and 9 in the morning. I’ve tried asking her this, but kids are shit at estimating time. “Just think Lena how long are you on the bus.” and I’ve sort of figured out that she walks her friends to class. But still?? is my kid just wandering around school for an hour in the morning? Do I even want to know?

Kids, the ultimate black boxes.