The struggle, the pain, finishing my masters last year. Memories now long forgotten. I swore never to do a PhD. I forgot. I started my PhD two months ago. I’ll spend the next three years analysing the transcriptome of cancer samples. Working at the Norwegian Radium Hospital. With an Illumina GA. I foresee much pain. And even more amazement. I was unable to resist joing the sequencing bandwagon, thankfully!
Russel Smithies (of AgResearch) suggested using Galaxy assisting in setting up a local GBrowse service (fetching UCSC Btau4 data, converting to a format compatible with GBrowse) at work (CIGENE, Centre for Integrative Genetics).
Galaxy is a web application that deserves attention! Fetching data, sorting data, combining data sets, alignments, manipulations, fetching other data resulting from manipulations, statistics, ++. All with a very-very nice, simplistic interface.
“It is designed to help two communities that often have difficulty communicating:
– Experimental biologists: “I really have no time to program but I want to do whole-genome analyses to find targets for experimental validation”.
– Computational biologists: “I develop algorithms but have no time to develop interfaces”.
There are also a number of screencasts available.
Example of usage:
“Suppose you want to find top hundred protein-coding exons in human genome with the highest density of single nucleotide polymorphisms (SNPs). Answering this question is not trivial. To do so one needs to compare all human exons to all human SNPs. To put this into perspective the current version of human genome (hg18) includes 357,517 known exons and 12,351,941 SNPs.”
Elegantly solved, using Galaxy, in this screencast.
Galaxy also lets you create workflows (screencast).
I’m very impressed, and I’m already planning a local Galaxy server integrating our local data sources and tools.
Now, all that’s needed is to teach people how to use Galaxy. Hm. And I probably have to look for a new job…
I actually did it. I defended my thesis last week. Believe it or not. I certainly do not. At least not yet. A summary of the thesis can be seen in the above picture (courtesy of Wordle).
Lately, being part of an academic environment, I have increasingly been aware of the insufficencies we face as human beings. Not that I don’t gape in astonishment of what is being achieved in both medical and technology related research advancements these days. But we have the potential to achieve so much more. And I feel, more than a little, annoyed by the way we seem to deliberatly waste resources on obviously counter productive projects. Productive projects that eventually would make, pardon the cliché, our lives better. Pouring billions of dollars into military operations is obviously an easy way to make friends with people running the military industry. But it’s not much affecting the way we deal with treating cancer, enhancing ways of expoiting alternative energy sources, speeding up the making of a space elevator, or understanding the complex ways of gene transcription interactions. And I do believe those extra billions could make an impact. However, thankfully, we are making progress nonetheless. And the progress is gaining speed in a way only J.B.S. Haldane and H.G. Wells could dream of (while Ray Kurzweil knew all along…). But it’s not enough. Professors and other scholars, with life long knowledge and experience, do retire. And even tho they are able to share their valuable knowledge through their companions and peers before leaving academy, much is lost. I want humanity to be able to pass on this knowledge in a new way. I want us to be able to share everything. Knowledge. Experience. Not in writings. Not in teachings. The process is too slow. But in a way numerous scifi story writers have been telling us for years. My new boss, Stig W. Omholt (director of Centre for Integrative Genetics), showed me not long ago an article in New Scientist. About researches involved in technologies “extending human capabilities“. We don’t have time waiting for evolution to make human brains inter-connect, sharing intelligence, knowledge and experience, vastly improving our scientific progress. We need to understand the way the human brain store information, and its nearly endless creativity. And be able to reproduce it. Thankfully, The Singularity Is Near. And it will change everything.
Rune started playing with Trixbox this week. Even got himself mexican and US based DIDs! So. As I’ve been planning on setting up an Asterisk SIP telephone server ever since I was inspired by Øystein Homelien when I worked for him at PowerTech sometime last century; Rune is to blame that I spent the last two days on doing exactly that. I signed up with a norwegian SIP provider, Phonzo, and got a nice deal on a Siemens Gigaset C450 IP DECT+SIP phone. And it works pretty well, it even has STUN support.
But. Why didn’t I know about ENUM before now?! I fell in love. Instantly. I enabled ENUM support on my Trixbox server, using Sipbroker, and registered my PSTN/GSM numbers at e164.org. Now, if only people could stop using that horrible Skype service (I refuse to link them), and start using services like Free World Dialup and Gizmo, and register the number in the ENUM database, we would all be able to call the the whole world for free. Regular telephone companies must be scared to death. No wonder the iPhone doesn’t have a SIP client – I bet Apple wants it, but I bet even more that AT&T refuses to accept it…
So. All of you: Get rid of Skype. Get a SIP (soft)phone. Get a free number. Register (ENUM) your current phone number. Make it point to your Internet SIP phone. Tell your friends.
Hm! As if I ever want to call people… (I still prefer e-mail for most communications. Good thing I registered my e-mail address with my ENUM entry!)
I recently bought an Apple Airport Extreme N. Mostly to get a more or less decent NAS. I was pleasantly surprised to see that it supports IPv6 out of the box; both automagical 6to4 and manual sit-type configured tunnels. As my trusty old IPv6-ISP PowerTech seems to have, temporarily (?), stopped caring about IPv6 (tunnel up, upstream IPv6 links down), I had to find an alternative tunnel broker. I ended up with SixXS. Got me a tunnel and a /48 net with reverse DNS to a low latency POP in Sweden. The tunnel terminates at a linux server (hosting this web page), from which I have delegated a /64 net to my Airport via a static sit tunnel. It’s working great, and all my hosting services are now multihomed. Not as great as if my physical upstream ISP (Lyse) would provide native IPv6, of course, but still. Now, if only you could nag your ISP to provide IPv6, maybe we could all finally get rid of that nasty NAT! (Oh, wait, that’s right, Apple was forced to firewall all IPv6 traffic by default after massive negative publicity. And there is no IPv6 NAT-PMP/uPNP yet. Here is an interesting read on the topic: Turning a feature into a flaw.)
Some months ago, my dad bought a Navibe GM720 (from Clas Ohlson, ~550NOK) to use with his laptop and AutoRoute while going on vacation to France. A couple of weeks ago, I started searching for an alternative to AutoRoute using my Mac. I found RouteBuddy. But I’m slightly turned off by paying $200 (program, maps for France and Scandinavia), when Autoroute is less than half that price. So I searched some more, and found Roadnav! It’s free, open source (GPL), and works with my Mac. Even the maps are free! For maps in Europe, Roadnav uses the OpenStreetMap database. Only … well, there is a high discrepancy between The World and OpenStreetMap, said gently. Some scattered roads, no street names, etc. Especially in places like Norway (excluding Oslo, the capital). Too bad. Well. Not really. I connected the GPS receiver to my MacBook, downloaded some software, and started driving around. GPSd while driving, to get the coordinates. GPSBabel to convert the raw file to GPX. And JOSM to map the coordinates into streets and upload to the OpenStreetMap project. I spent this evening listening to my favourite cellist, Truls Mørk, drinking whisky, and mapping parts of Fredrikstad and Ås. Not at all bad. Now, if I only spend the next 200 years driving, I might cover a fraction of Europe, to make the maps I thought I needed…
I recently convinced my sister to buy my beloved MacBook Core Duo, while I upgraded to the MacBook Core 2 Duo. Being lazy, I just swapped the hard drive from my old CD to the new C2D. I ran into two problems on my C2D:
- 1: The trackpad was recognized only as a mouse. I was unable to use all the life saving features like “two finger tapping means right click”, and “hold one finger, use the other to scroll”.
- 2: The function keys didn’t work. (To manipulate screen brightness and audio volume.)
Seaching the net revealed no clues. Except a linux kernel patch. It turns out the CD uses the Geyser III trackpad, while C2D uses Geyser IV. I dived into this, for me, rather unknown Mach/OS X-territory (altho I consider myself a long time unix zealot), and located the file /System/Library/Extensions/AppleUSBTopCase.kext/
Contents/PlugIns/AppleUSBTrackpad.kext/Contents/Info.plist. My old hard drive, from CD, contained no information of Geyser IV. Stupidly trying to edit this manually, and reloading the kernel extention (using kextunload and kextload) was, of course, unsuccessfull. But, behold!, copying the AppleUSBTopCase.kext-directory from the new hard drive fixed everything. I am now happily scrolling using a new MacBook C2D using the original OS X installation from my old MacBook CD.
VirtueDesktops, Sidenote, Adium, Google Notifier, Connect360, Growl, RCDefaultApp, Flip4Mac, VLC, Journler, Cyberduck, iTerm, Mail.appetizer, Shrook, Xtorrent, Corripio, 4Peaks, EnzymeX, Geneious, Coot on OS X, NeoOffice, Firefox, LiquidCD, Quicksilver
Our lab environment is polluted by not only cancer inducing chemicals like ethidium bromide and the neurotoxic polyacrylamide. Even more disturbing is the frequent usuage of the radio. Or, to be more precise, the lack of a radio station that does not suck. Here in Norway we have a couple of commercial nation wide channels in addition to some channels provided by the public broadcasting institution. I can’t stand commercial channels, with the non stoppable annoying advertisements. But, even worse: The channels we listen to (we switch a lot) have one thing in common that annoys me the most. The repetition rate of some of the music. I mean, I don’t want to listen to the same tune 9 times during just a few hours.
Lately I have started wearing my trusty old DAP, which luckily carries an endless number of hours of music. After a few weeks I have come to the conclusion that I prefer some kind of jazz inspired music while doing lab work. Even tho my favourite band this week has been Gotan Project, which is a fusion of tango and electronica.
(Yup, Nat King Cole works pretty darn good for me.)