My lab's aging LaserJet 1200 recently started failing in an interesting fashion. It would print a page nicely, but then attempt to feed a second page before the first had finished, leading to a jam. Some googling lead me to this page which indicated that the feed solenoid was sticking. After a failed attempt to clean gunk off of the feed solenoid arm (and failing to find anything else wrong, I wrapped the feed solenoid arm in scotch tape and attempted again. Viola, a printer working once again.

Updating the bios on my X200 Tablet was surprisingly simple in Debian. Often, updating bios requires some sort of Windows or Dos boot disk and all kinds of machinations to make that happen. Lenovo provides an iso image to update the bios which alleviates some of these problems, but as the X series don't have built-in CDROM drives (and I don't have an ultrabay),

All you have to do is download the iso, install syslinux aptitude install syslinux, and configure grub to start memdisk and boot off of the iso. You can do this by copying the iso to your boot partition (cp 7wuj43uc.iso /boot/;), copying memdisk to the boot partition (cp /usr/lib/syslinux/memdisk /boot/;) and adding an entry to your grub.cfg like the following (putting it in /etc/grub.d/40_custom is probably the most reasonable):

menuentry "Thinkpad x200 BIOS Upgrade to 3.19" {
       insmod part_msdos
       insmod ext2
       set root='(hd0,msdos1)'
       linux16 /memdisk iso
       initrd16 /7wuj43uc.iso
}

Note that your configuration will be slightly different depending on whether you have a separate /boot partition (I do) or not.

Then you just need to boot the laptop, select the bios upgrade option in grub, and tell the program that you really do want to upgrade the bios and that you really do have a full battery and the laptop plugged into an AC adapter.

The major idea for this method came from ThinkWiki's entry.

If you're doing genetics and use SNP, you've probably often wanted to extract information from dbSNP. A long time ago, I wrote some scripts to scrape their webpages for information. Now that I'm dealing with millions of SNPs, that's no longer possible. I needed my own mirror of dbSNP. I did this as of build 132, and since they've now released build 135, I wrote up what to do, and shared the git repository which has the utilities and schemas I used.

This year has been fairly good for our lab; a few years ago we found an association of rs17849502 with SLE. Finally, after significant work by our collaborators, we managed to demonstrate a functional significance of this SNP with possible importance for SLE. Changing H389 to Q reduces the function of the NADPH oxidase complex by 50% in Vav dependent Fcγ response.

You can read the paper "Lupus-associated causal mutation in neutrophil cytosolic factor 2 (NCF2) brings unique insights to the structure and function of NADPH oxidase." (pmid: 22203994)