Wednesday, April 08, 2009

Recurse()

Assignment 8.2
Question

"

A typical neuron needs ~100 nearly simultaneous inputs to trigger an action potential. It connects to about 10,000 other neurons. Why might this situation lead to epilepsy? How does brain circuity avoid it?
Hint: Let x% of neurons be active, in a network with y% probability that any given neuron connects to any other given neuron.

Answer: By trying to actually answer this question. By settling for a qualitative (gyan) solution.

Tuesday, April 07, 2009

Dream in the Shell

Saw the movie Ghost in the Shell. First one followed by its sequel. Wouldn't it be really cool to have some test of physical reality? Just a thought.

Wednesday, March 04, 2009

Sunday, March 01, 2009

import Geeksta.*;

Scene: A mol. bio. gurl is bent over cutting out a gel to make it 'permanent', for scanning later on.

Mol. Bio. guy: (glances down from his high bench) Lets make a double helix.

Mol. Bio. gurl: Lets restrict it to just a simulation.

Mol. Bio. guy: My place.

Saturday, February 28, 2009

Royal stag nacht

Yesterday night was crazy.

Tuesday, January 06, 2009

Wednesday, November 19, 2008

Cheers

Lecture halls too obey the Beer-Lambert Law.

Saturday, October 25, 2008

Time to put gimp on windows. Will now boot ubuntu only after its newest release. Hate doing it though. Gimping is such a debian thing.

Friday, October 10, 2008

Inbox

Frequency of checking your mail = ( joblessness ) + C

Monday, September 01, 2008

Scene III
















CS has its own uses.

Behind the scene:
Kir2.1
Yankee
PT
Duds

P.S. Grrr...

Friday, August 22, 2008

Nasty over Curly of Oster

Arrogance of a fruitfly worker is a trait determined by the genotype of D. melanogaster.

Culture Curry

Cultured neurons is maggi.

Wednesday, August 20, 2008

Fly Interrupted: Scene II

Act I

A balding biochemist in his thirties. Lab. door shut behind him. Descends down the stairs. A racquet in his hand. A fellow worker asks about his bike keys...


Biochemist: Shit man! I've to put teflon coating on the bike. But my cloning did not work.
Fellow: You don't deserve a teflon coating.
Biochemist: Hmmm

Both walk away from the institute building heading for the badminton court while they praise the b'lore weather.

Tuesday, August 19, 2008

abbr. lol: Laughs Out Loud

How many times do you actually lol when you lol ?

Sunday, August 17, 2008

Chick Flight

A pretty chick will always be found accompanied by another hot chick. This is to minimize drag and maximize lift.

P.S.: Sympotein hangover.

Thursday, August 07, 2008

Boomer

Yet another fly lab scene:

Act 1:

ElSudCampeador: (chewing gum) Hey webrat want some gum?
Webrat: (euphoric) Wow! Chewing gum. Sure. (gleefully accepts, starts chewing).
Chaos: What man (grunts), you will wow at anything. Even chewing gum! (evil laughter)
Webrat: Well not at every kind of chewing gum; for instance me wont wow at glia.
Chaos: (dumbfounded) Hee. (back to his photoshopping, brain 101. Glial screen)

Webrat and ElSud burst into laughter.

P.S.: ElSudcampeador later asserted that he too had CG after a long time. Unlike Scotty, a chewing gum person.

Monday, July 14, 2008

Bumper

Even an atomic nucleus is not really smooth when you get to know him closely.

Sunday, June 29, 2008

First Law

If it is in the fly it will be solved.

Saturday, June 28, 2008

Tuesday, June 17, 2008

Whiskers

I wish the mouse had balancers.
I wish mouse could be made in 10 days.
I wish mouse had larval stages.
I wish it hatched out of eggs.
And there was fly!

Friday, June 06, 2008

Troll

Spare Drosophila melanogaster from pull downs. It doesn't really enjoy Westerns either. Yeasts love biochemistry.

Tuesday, June 03, 2008

Saturday, May 31, 2008

The D Bug

Fly brain was probably written in PERL.
Gal4>UAS:{ChR2/TntG} is python.

Tuesday, May 27, 2008

Lift off

Just reached Mars. The pole.

Sunday, May 25, 2008

Thursday, May 22, 2008

Screen

When you get your camcorder along flies refuse to show phenotype. Flies are camera shy

Monday, May 19, 2008

UT

And UT is UT.

Thursday, May 15, 2008

Shuttle

Singles badminton and doubles badminton are two different games.

Sunday, May 11, 2008

Blurrrr

Motion_detect() == Edge_detect() in time.

Friday, May 02, 2008

Stinky

Do you remember the smell when we fell in love?
Do you remember the smell?
Now baby
Do you, Do you
...

Wednesday, April 30, 2008

Free Will

Free will is only as free as the neurons allow.

Saturday, April 26, 2008

Developmental Determinacy

How much breathing space does the fly genome allow it's neural system?

Wednesday, April 23, 2008

Fly-Will

Is a fly's free will all about the indeterminacy emerging from the complexity of its neural connectome?

Fly-Walk

A wingless fly. Starved. Food odor. Watch fly-path.

Would this fly have taken a different trajectory in a parallel universe?
How many parallel universes to account for entire trajectory space?

Tuesday, April 22, 2008

SimpleXMLRPCServer

Just wrote* a command-line based ping, mssg, picture send/recv system which:
  • is Multi-threaded
  • uses XML based messaging
  • uses RPC
Using Python's SimpleXMLRPCServer (shipped with the standard distribution) in under an hour's time. Python simply kicks ass when it comes to speed of development. Also its so much easy to think in python.

Plan to add voice messaging ability and encryption support to it. Tomorrow. Day after.

Anybody else want some!


*I use Vim. Vim is sweet.

Sunday, April 20, 2008

Wanda the Phish

The close resemblance of the recent posts to Wanda the Fish Fortune teller is purely coincidental.

JLT

Wanted to write something profound. But forgot what.

Saturday, March 15, 2008

Conversation in a fly lab

Guy A: Do you have CS?
Guy B: You mean Canton S or Counter Strike?
Guy A: Photoshop CS.

Thursday, March 13, 2008

Catalase

Catalogues received at conferences shall be properly utilized. As RNAse free paper.

Sunday, March 09, 2008

P value

Its been happening a lot to me recently. Everytime I look at my watch it shows some interesting time. 22:22:44 or 5:55:55 or you know of the sorts. Always (almost) some pattern in it every time I check.

Saw, what I think, is a shooting star yesterday night on the way back home from the lab. Pretty fat white blazing mass with a bluish hint in its tail. Bigger than I have ever witnessed before.

And today. Read some book by the pond side. Book reading at its best. Life of Pi, amidst a garden, on a metal bench by the pond side. That was some experience.

Saw pre-sunset sun stuck behind tall green vegetation. Orange disc interrupted by spinach green tree tops. Either the time is good and I am lucky or my current environs are so much abound with beauty that wherever and whenever one looks one gets hold of some pleasing sight.

Friday, February 22, 2008

Fink before you leap - PIL Installation howto for a Mac


PIL (Python Imaging Library) habit is difficult to let go. And well a Mac is a Mac. Follow these steps to get PIL up and running on your Mac.

  1. Install fink. fink = an apt-get like (ditto?) package manager for your Mac. apt-get also gets installed along with it. Just google for fink and you will find its dmg image on sourceforge. Installation is simple point and click.

  2. Configure fink by issuing the command
    fink configure
    It will ask you couple of easy questions. Bear with the thing. If you use a proxy to connect to the internet its necessary to specify it during this configuration procedure. *IMPORTANT*: Write fully qualified name of the proxy (with the protocol). For e.g.
    http://proxy.ncbs.res.in
    I had dropped the protocol during my initial configuration and fink failed to connect to internet. I wish fink architects had arranged so that the proxy config defaults to http protocol. Newayz. Now you know!


  3. Then issue this command
    apt-get install pil-py23
    This command will install all the necessary dependencies along with PIL. May take a while depending on your net connection speed.

  4. Test your installation by
    python2.3
    import Image

    P.S.: Looks like PIL is currently available only for python2.3. You might wanna consider symlinking python2.3==>python in your /usr/bin. I am undecided for my Mac mini shipped with a python2.5. And I hate putting that extra "2.3" to launch the interpreter. What life !

    P.P.S: Thank me. And then thank Raj who egged me on to blog this.

Monday, December 31, 2007

tumb tumb...

So much to write home about. Not really up for it though after a new year's bash ;)
Anticipation is all time high and the scale of happenings is going to go up exponentially with each day that passes.
Nervousness, anxiety, adventure, excitement and sheer joy. A mocktail of a different kind to give one the high of highs.

Sunday, November 18, 2007

Differentiated

Behavioural Neuroscience it is.
Making fruit fly larvae race and flies jump is gonna be the way of life. Of course with a little bit of genetics and electrophysiology and imaging and biochemistry. And perhaps some gene construct designing and transgenics :D *high, delirious*

Dunno if its gonna be Mumbai or Bengaluru though.

Saturday, October 06, 2007

The Chain of Chains

An E. coli cell divides. That the DNA is passed on to every "new born cell" post-division is no big deal. But along with DNA the daughter cell must also receive at least a bare minimum non-DNA things to make use of this genetic information. Which is why a cell division involves distribution of cytoplasmic goodies between daughter cells than just budding off of a replica of the original DNA molecule.
This means that every cell, each and every one of them that we are made up of and those we see around us in other beings have at least once shared a single cell membrane at some time point since the first membranated organism in the primordial soup upto this moment. And this truth will remain so as long as there continue to be cells. The mother of all chains - this cytoplasmic connection.

Tuesday, September 25, 2007

Fly Custards and Fly Jelly

Fly Custard
The corn medium has smell. I wonder how much influence that smell is having on the larval olfactory circuit which I intend to put to "adaptation/recovery" tests. The smell is a cocktail. And my worms have ploughed through it endlessly before they were abducted for a larval plate test.

Can the smell be fixed? So lets think of something else. Wish there was a Matrix like "pod" for the larva. A larva living its life oblivious to "reality". Life would be so easy for the fly worker. el yo el

Excuse me on my lack of information as basic as this. I guess the fly cannot be grown on just glucose. The vitamins et al things must be imported into the worm body as it develops and the yolk resources get exhausted. But m gonna test this one.

Fly Jelly
Lets get a fly couple to mate on an agarose gel block soaked in glucose solution. A sort of fly date on the beach el yo el. Then see if the eggs hatch and observe how far a larva goes through its development. This way there would be little smell that the larva would have seen in its "past life" and would be an olfactory virgin when I show it the aromatics during my plate tests.

P.S.: A genetic way of doing it (one of the ways (at least in principle) ) would be making a line which is conditional mutant for olfactory receptor expression and induce the receptor expression just before your plate test. But I suppose it could also have its caveats and may be plain un-doable.

P.P.S.: The P.S. was put there to save face just in case the glucose-gel-beach way was found to be just too ultra low tech by some. :D

Sunday, September 16, 2007

On Evolution

A gene may get duplicated. This may be followed by mutation in one of the copies. Functionally if this "mutated" version is very different from its counterpart then one will call it a new gene rather than simply an allele of the "original" gene. Thus a species gets a new gene. But it may also simply acquire a gene "horizontally". The rates of both these phenomenon taking place would be expected to be determined by the species under consideration. Still one may ask, generally how do they compare in their contribution to creation of new genes?
Furthermore, once a gene has been acquired by a species its maintenance would be contingent upon its contribution to the species fitness. It would be conserved if it is useful else lost if redundant (or found to be deleterious). What are the rates of this "gene dumping" phenomenon? Wondering if these "rates" could be quantitated. Then it would be interesting to compare the rate of acquiring a gene and the rate of losing a gene. Dont remember why I thought of these questions but still it would be fun to know their answers if possible.

Friday, September 07, 2007

Wiring - An Interlude : The 2nd Rotation


The promised wiring blog posts are still to arrive. Maybe I am still ruminating the ideas expressed in the last two neuro lectures (its been 2 weeks already). One was on olfactory and the other on optical path-finding. Who knows, these posts may not even happen.

Newayz.

The olfaction lecture + couple of posters on olfaction + AWS + fly seems like such a cool system + other things = too difficult to resist rotating in the olfaction lab.

So here I am. Begins officially this Monday. Already been assigned to a "project" and given a dozen papers to read. The latter in preparation of a sort of a mega-journal club this Tuesday.

If you smell what the rock... :D

P.S.: Picture - C155-Chops larval pre-exposure to excitatory light just before the the "Larval Plate Test".

Wednesday, September 05, 2007

Wiring : An Interlude - DBS TIFR AWS On Adult Neurogenesis

Epigenetic mechanisms speculated in neurological stress response in maternal separation paradigm of mice neurobiology model.

This one hit me like a bullet. Shaken.

Sunday, August 26, 2007

Wiring : An Interlude

The answers are coming.

-Morpheus to Neo, The Matrix.

Friday, August 24, 2007

On Biological Evolution

Neo-darwinism is a thing of beauty. These days it hardly happens that one does not suspect it to be behind a given biological phenomenon. But it mostly comes as an after-thought. I wonder if it has any predictive use at all.

Monday, August 13, 2007

Wiring 1

Was feeling a little under the weather. I hope a dengue bearing mosquito is not to blame. Damn! I could feel I was not at my analytical best. But it was a neuro lecture, so a hightened sense of understanding was my default state. Could have been much better.

So the questions...

The lecture moved on. I was getting kind of desperate as it seemed the answers weren't on the menu. The lecture began with a recap session which seemed unusually long. Then the flow of the discussion kept getting led astray mostly due to certain over-enthusiatic individuals who seemed to get ahead of themselves. A pack I think I once certainly belonged to and one which I think I am transiting. Let the prof talk, you might lose out on some vital insight in your attempt to push/validate your ideas amidst the discourse. This is some sort of a policy that I have (unknowingly?) adopted these days. Wait and watch and pounce when you just have to, an almost compulsive act.

All in all kind of less than my expectations (greed?). But...

At the end along came an idea. A classification of memories was touched upon slightly. Motor memory and declarative memory. My face might have lit up. Yeah, just when I was gonna conclude a wasted lecture. "Functional memories". I felt reassured. I have not been reading useless stuff after all; amidst short naps on the D-block library couch.

I am glad I saw a thread. At the end of the lecture. So be it. Have been promised a talk on the developmental problem of the sense of smell in the next lecture by a seasoned and a highly motivated student. Also have been urged to do some home-work on how the smell problem was such a special problem before it was cracked by a certain Linda Buck. Will fit in the home-work thing tomm since kinda jobless these few days. At least until 21st when I get my time on the NMR machine again...

The sense-of-smell talk is going to be interesting in every way.

Friday, August 10, 2007

Reality Check

After attending a neuro causerie.
I wonder whether when I myself master the jargon and fluorescent imagery my past scriblings will seem even to me as nothing more than idle doodling. (Not that they don't at all seem so now :P)

Thursday, August 09, 2007

The Wait


In the coming couple of weeks my neurobiology course will conclude. It will end with how-the-brain-gets-wired themed lectures. I guess two of these plus a lecture focused on a special kind of wiring - the wiring of the optical pathway. Nothing less than a feast.

Following are the typical questions whose answers I will be fishing for in these lectures.
  • The basis of instincts. Especially the behavior of just born calves. Where does this information come from?
  • Nature of data structures. Is the location and form of a given memory identical (or at least similar) in two different humans? For e.g. say the neuronal data structure corresponding to a given memory - say the English numeral 1 (i.e. basically a shape).
  • Imagination. Imagination is a brain activity unlike other brain activities such that you have neuronal activity in absence of an external input. How does one imagine?
  • Perception. Making an image of the world around you is a biological challenge in itself. A 100% biological lens followed by a bunch of 100% biological cables and ultimately a 100% biological "processing center". Marvelous. Still, beyond that how does an individual perceive an object. For e.g. when I see a square I feel a square. I wonder if some kind of reflection phenomenon - reflection of neural impulse encoded information within the brain would be necessary for perception. Say back and forth between the two cerebral hemispheres. This question is perhaps intricately related to the puzzle of imagination stated before.
Yeah, pretty much it. I hope to get at least a start towards solving these puzzles from these lectures. A reasonable expectation I suppose.

Wednesday, August 08, 2007

BioSphere


Biology is like a sphere. Every worker thinks the field he is researching is central to biology.

Monday, August 06, 2007

On Biological Evolution

Following two general ideas about biological evolution are the result of an informal discussion with a fellow colleague.

  • The period just before advent of cell membranes is what I shall refer to as the late primordial soup times. Assuming that during this period organic compounds which would be sources of carbon skeleton and free energy would be available in plenty and hence would not be a limiting factor for the survival of contemporary biologicals; it would be a free lunch. One would therefore expect the evolutionary pressures to be relaxed. However, after membranes were invented these diffuse biologicals would get a sense of individuality. Evolution would now have a well defined target for it to act upon.
  • When faced with a challenge only those cells who have invented a biochemical pathway to solve it will survive. But it probably was when the cell found a way to document this information in the form of genetic information that evolution got a firm substrate to work upon. With it also came the concept of "next generation of cells" who would re-use their ancestral biochemical logic than re-invent it. But the beauty is that the system still allowed for new experiments, in essence allowing the ancestral logic to be overriden.

The events following post-genetic material times probably are just biology trying to solve its own engineering problems.

Saturday, July 28, 2007

Functional Networks

Was reading a book in the D-block library today. Some brain-memory book, will tell you the name next time I edit this blog.
A pretty fantastic idea was expressed in it or so I believe. That it would be wrong to think of brain memory on the lines of computer based memory. Computer based memory has an "address" element attached to it. But a brain "memory" is functional.
By functional I mean that say in response to an external stimulus an individual has say two different reactions and the reactions are such that they require the usage of motor neurons. Each reaction will have a motor neuron firing pattern. The decision as to which of these two motor neuron bunches is fired will rest upon input from common node or a branch point. The memory read operation is thus integrated with its actual implementation or the final outcome.
The decision as to which bunch gets fired at a given time is clearly determined by the strength of the connection to this node. Beauty is this strength of connection is experience modifiable.
But hey, it was just a model...experiments might show an extremely strange picture. Plus this is a Saturday evening blog post :D
You never know.

Wednesday, July 18, 2007

Sab maya hai ?

virtually real ?...maybe. *shrugs*

Saturday, July 14, 2007

Fixity



...that one always has a choice.

Wednesday, July 04, 2007

3 . c 0 L 1 1%

>gi49175990refNC_000913.2 Escherichia coli K12, complete genome *


. . .

I guess I didnt have anything else to post. Sometime back I downloaded the E.coli genome sequence and placed the fasta file on my desktop for my amusement. Whenever I looked at that file I could not help but smile in my mind and nod slightly at the screen. Here is an organism - a bacterium frozen in a text document. Life cannot get more digital...



*only 1% sequence was posted, posting the whole genome would freeze the browser.

Monday, May 28, 2007

...phew !

but this is only the beginning...

a sense of belonging...what one does now is no longer going to be just personal curiosity...a concerted, directed effort backed by dedicated personnel and resources.

no trivial pursuits...not that they were ever so to oneself...but the way the world sees it...

one has now been absolved of the guilty pangs accompanying post-Lewin et al sessions...

i had coffee twice today. one frappe at the bangalore airport. one hot coffee aboard the Mumbai bound flight...

ones cAMP must be on overdrive now...caffeine can drive one crazy...phew !

Wednesday, May 16, 2007

Something On Genome Wide Repeats


Was readin up on chromosome organisation in Watson. Something came to my mind. Two things actually.

Transposon Dating: How rare are human transposition events? Can those sequences, which are relics of transposition events found in the genome-wide-repeat subset of the intergenetic DNA, be used for phylogenetic analysis ?

Gene Pool Hypothesis: DNA sequences in the genome wide repeat subset of the intergenetic zone are thought to be relics of transposition events. They play no functional role in the species with respect to lending their DNA as an expression template. Can one expect the evolutionary pressures to be relaxed on these zones relative to actual protein coding/regulatory genetic zones? If so, then could this rich pool of variable DNA sequences serve as a birth place of novel genes. A genetic nebula. Such novel genes, if they exist at all and are discovered would be expected to bear little homology to other pre-existing genes.

If a tree based on sequence homology for all the 27000 human genes is constructed what would it look like?
1.What is the kind of homology amongst these 27000 genes?

2.Out of them how many would have a true common single ancestral gene and how many would have formed independently of this ancestral root? Or how many roots can account for the entire human gene repertoire?

3.Would these independently formed nodes in this tree be good places to look for genes which makes humans more so than say other primates?

4.Have any of the above independently formed genes been derived from mutation in the transposition-events relics found in the genome-wide-repeat zones in the intergenetic DNA (discussed previously) ?

5.Can the above mentioned zones give rise to sequences coding for trans-regulatory-RNA apart from or if not full-fledged genes?

If any of the above 2 mechanisms are real then some of the complexity of humans can be ascribed to the above gene forming pools of the junk DNA component of the human genome.

Friday, December 01, 2006

Sun Java Application Server - MySQL : Connection Pooling HOWTO

Once in a while I come up with some tech howto. The last one being "HOWTO set XMMS current song as you IM status".

This time it is...


MySQL Connection Pooling Step-by-Step HOWTO

  • Make sure U have the mysql.jar file with U. It contains the mysql driver.
  • Your Sun Java Application Server (SJAS) folder looks like this

drive:\Sun\AppServer\ i will refer to this location as J2EE_HOME

  • Now copy the mysql.jar to J2EE_HOME\ant\lib directory
  • Start the server
  • Open the admin consolehttp://localhost:4848/
  • Login using the default uname/passwd

uname: admin

passwd: adminadmin

  • Click on : Application Server => JVM Settings tab => Path Settings tab
  • Now paste the following line: ${com.sun.aas.antLib}/mysql.jar in the textarea labeled as "Classpath Suffix"
  • Click on Save. Close the admin console.
  • Restart the server.
  • Open the admin console (http://localhost:4848/) again.
  • Expand the Resources node
  • Expand the JDBC sub-node
  • Click on Connection Pools
  • Click on New
  • Fill up the fields as follows:

Name: mysqlpool

Datasource Classname: com.mysql.jdbc.jdbc2.optional.MysqlDataSource

Resource Type: javax.sql.DataSource

Do not modify rest of the fields.

  • Now add following properties by clicking on "Add Property" at the bottom of the form:

DatabaseName : OnlineBankingSystem

User : admin (your mysql username)

Password : admin (your mysql password)

  • Click on save. You will see "New values successfully saved"
  • Now, Click on Ping. If you get a "Ping succeeded." then be happy :D
  • Now is the last step. click on the JDBC Resources node.
  • Click on New.
  • Fill up the fields as follows:

JNDI Name: mysql

Pool Name: mysqlpool

  • Click on Save. Thats it! We r done ! :D
  • Your mysql connection pool is now ready to accept requests to the database. It is referred to as just "mysql" from within the EJB and JSP (its JNDI Name remember, :-? ).
  • And of course, I wish U all the Best ! :D
    Lemme know soon wat happened and if it helped. :D

Sunday, September 03, 2006

What does a CD mean to a cow?


Passing thought: Imagine two people conversing using their Nokias. The information is ferried between them through space in form of light waves. Would it be possible to monitor the physical properties of the space and infer the information being exchanged? One may point out that the cell phone data is garbled for obvious reasons. But would it be possible to determine atleast this garbled data by performing a suitable measurement on the space, somehow?

Eavesdropping on people's conversations is not really the topic. It is rather about the nature of information.

Is there a theoretical limit to how much information a given volume of space may possibly
contain irrespective of the method used for storing/encoding this information? There can be two possibilies:

  1. Yes, there is a limit.
  2. No, there isn't a limit.
If it turns out that the first case is true, then it would imply that there must be something in the fabric of this universe - some key ingredient which serves as a universal currency of information. This would mean that a general purpose instrument which can quantitate information contained in the given object space should be achievable.
But if it happens that the second case is true then information would turn out to be indeed something very personal - so that each individual can have the freedom of interpreting this information the way way he pleases. That there wont be such thing as a universally true information or a universally false information for a given obeject space under scrutiny. In such a case it would not be possible to develop a generic tool or quantitating the information content Which is :-( So, clearly in this case infomation exchange is possible only if the two parties agree on the information content.

Passing thought: Your Win-Zip program which you use to compress data does its thing like this[2]. A CD contains data stored along the tracks. In each track there are two possible features - a small pit and no pit corressponding 0 and 1 or otherwise. A 700Mb of such a CD would contain an entire DivX of The Da vinci Code. One would be able to watch it if one has a CD-ROM to decode the engraved data. But what if one does not have such a specific device? Does the information contained in this CD affect the properties of the physical space it occupies so that a suitable measurement on this space would reveal the movie data indirectly?

Casual Mention: "Biology is the domain of the complex. It takes 3 * 10.pow(9) bases = 6 * 10.pow(9) bits of information to specify the DNA that determines a human being".[1]

But what is in it for the biologist? Well, putting it rather (very) bluntly,
If the information is found to be universal as opposed to personal then a given space, say a space which contains only a single DNA fragment, can be analysed by using some tool to come up with the absolute information content of the space; DNA sequence in this case. Outrageous. I know.

Theres more actually, but by now I have either forgotten to mention those things/ too lazy to type anymore/ its lunch time :P
Maybe some other day. . .

Interesting reads:

  1. http://www.cs.auckland.ac.nz/CDMTCS/chaitin/***
  2. http://en.wikipedia.org/wiki/Data_compression
  3. http://www.faqs.org/faqs/compression-faq/part2/section-4.html

*** = Must read.

Wednesday, June 07, 2006

T.I.F.R. BioChem Labs


Plush, clean, efficient publishing machines, silent except for the ho-humming of speed-vac rotors or gel-shakers amidst over-crowded chemical shelves with a graffiti of TODO post-its whereas the room corners are claimed by the latest and greatest fat gizmos spitting out raw data only to be handed over to the computing room to make sense of all that alphabet soup while the show is run by smart biochemical ensembles of CHONSP.

ΔG = ?

Tuesday, January 17, 2006

Million Dollar Meccano



Some Basics
Everybody knows the central dogma of molecular biology - DNA=>RNA=>Protein. There is another lesser know dogma - protein primary sequence=>protein 3D structure=>protein function. The real challenge now is to actually be able to predict a protein's function given its amino acid sequence.
Proteins come in mind boggling range of shapes and functions. Yet all of these exotic stuctures derive their blueprints from a modest (boring), 1D stretch of information encoded in a mRNA fragment. Hence, essentially, manufacturing a protein is simply (!) the folding up of a linear polymer of amino acids onto itself through twists and turns and coils assuming a complex 3D structure which can do useful things.
There is considerable evidence suggesting that the protein's primary structure contains sufficient information to direct the folding of the polypeptide backbone to give a correctly folded 3D sturcture.

Show Me My Millions
Biology never had it so good. We now seem to know much of the basics, we seem to have the suitable technology and most importantly we also have brains working towards this massive goal. It is just a matter of who claims the top cherry. Biology is therefore going to be an increasingly interesting area even for the VCs. The X Pize Foundation of the first-private-spaceship-contest fame is now planning to offer a new prize '$5 million to $20 million for anyone who can decode the DNA of 100 or more people in a matter of weeks'. This gives one an idea of how important the once academic questions of biology have become relevant to the general masses. What would the next prize be . . . protein folding?
These are not the millions I was talkin about. These are just the tip of an ice-berg. By putting together 20 amino acids together biology has produced a plethora of protein structures and this was pretty much to be expected given the diversity that surrouds us. But even these different proteins are made of a large but limited number of motifs. Once all these motifs are charaterised one would truly have the nuts and bolts out of which actual functional proteins - molecular machines could be made. This would open up an entire industry of protein engineering. A good part of which may focus on chemically tweaking the existing proteins and a maybe a smaller group may focus on inventing entirely novel proteins. The possibilities seem to be endless. People are already using biological macromolecules in household products - e.g. Subtilisin a proteolytic enzyme is used in these.

I can go on and on . . . got the point ?

Wednesday, January 11, 2006

IM Gaim

Of men and women and there IMs...









Why am I rambling about some such topic? Is it really unlike of me to jabber on non-tech topics... :-? ewwwwwwwww lets explore the not-so-tech side...of people's IM-habits, IM-eccentricities, IM-egos, IM-hyphenated things.....

Lets start with a Sunday morning newspaper MCQ
Quantitating ur IQ: IM Quotient

1. Buddies Count?
a. Who counts...c'mmon
b. ummmmm
c. The messenger length/breadth is higly distorted (lengthwise)

2. Buddy Online/Offline ratio?
a. very very less than 1
b. approx 1
c. much greater than 1

3. Who's the 'initiator'?
a. me...i am a very good guy/gal ya know
b. depends
c. how do u do that? every time I login IM screens mysteriously pop up everywhere

4. Ur IM input/output ratio:
a. I am gr8 at typing ya know...my im windows are splattered with my say
b. Neva thot bout that really
c. **shrugs**

5. Away message
a. Who's away? me ...r u kiddin me
b. Sorry, I ran out for a bit! (classic Gaim)
c. i did'nt know ya miss mi so mach laddie

Ok lets see how U scored:
Mostly As: Its CyberSpace in case u did'nt notice...u can do without being urself..."Wake Up"
Mostly Bs: Well U r managing it well fella.
Mostly Cs: Ok...you dont have to be truthful evertime

After that taxonomy of IM creatures lets chk out something more about the IM-culture.

Dogs and bitches...
Hyenas, spiders, bees, dogs and other of those NAT GEO creatures. All these species have one common characteristic - the male must make the first move. Humans stick to their animal instincts in the IM land too. When a female collegue/ friend/ UFOD (undefined-female-object-of-desire)/...sorts... come online it is the moral responsibility of the male to 'initiate'. Female initiations are kinda rare cosmological events and need careful investigation.
Cant we just leave behind all our inhibitions when we are atleast IMing?

Should I/ Shouldn't I moments...
Okay so u login to ur account. U see this list of online buddies. U r not exactly in a mood to chit-chat. Okay with some of them, but not others. It is considered to be rude if you r online and donot IM with other 'available' onlines. So wat u end up doin is waste some bandwidth with some non-sense for 2 mins. Hang up with regular 'tc's and 'bbye's and ...the sorts. Why do all that? If u wanna chat plz initiate, if they want let them do it.

Bloated Egos...
Giving out ur IM id to ur boss is a very BAD idea. Bosses expect u to initiate, to do most of the chattering, entertain them...Besides the bosses, some other people around you who are'nt exactly ur bosses but create such an impression are like pseudo-bosses. These are the domineering sorta people. They have a kind of air to themselves. They will reply least frequently, with minimalist use of words, with huge latencies.

The END Problem...
Okay...this is not the end problem of molecular biology (d-uhh DNA replication)....this is a classic IM end problem. Let me put it straight - "How does one end an IM conversation?" People have this habit of trying to end the convo by these one liners:
tc
gnite
bbye
cya
ciao
chao
c u around
lets part
over n out
. . .
But no matter what u type...the person on the other side has to come up with some reply so that he-may-have-the-last-say.
The Solution: Well...it seems the problem has been solved by biology. When u wanna end a convo just type UGA (or UAA or UAG). WHY? Well go n revise ur high school biology. UAA, UGA n UAG are stop codons. What bout the non-science people U ask? In case U did'nt notice this is basically a tech blog and the intended audience is obviously other techies/nerds/geeks. What bout non-bio-science-people U ask? Well... if U r a non-bio-science-fella ; I hate to be the 1 to break it to ya fella but "U dont fit into the l33t group, aadeos" UGA.

more, comming soon...

Sunday, December 04, 2005

Appreciating Randomness...E.coli Interupted

[mycrap]
Presenting a classic problem in genetics...a great way to introduce oneself to Mr. Poisson (of the distribution fame)...the joy of modelling...
[/mycrap]

Problem: The transfer of DNA from Hfr begins at oriT. The gene closest to oriT on the donor chromosome enters the recepient cell, followed by successive genes on the donor chromosome. The frequency of a particular gene entering the recepient is a function of its distance from oriT. Moreover, this frequency is found to decrease "exponentially" with a gene's distance from oriT. Why is this decrease exponential (and not linear, say)?

The Solution:
Poisson says
"Let m = the mean number of successful events. The probability of k successful events is given by the formula -
P(k) = (e^-m * m^k) / k!"


The key to solving the given problem is to define the 'm' value. Let us define the m value as the 'number of breaks per unit length of the chromosome'. So if 'x' is the length of a given chromosome segment the number of breaks in x will equal 'mx'. Remember that mx is the average number of breaks in a population of chromosome segments of length x. That is some x-lengthers will have more that mx breaks and some less than mx and some may even have zero breaks.

Even a single break in a length x is sufficient to ensure that a marker at distance x from oriT is not transferred. In other words any non-zero number of breaks in x will prevent marker at x from being transferred.

Now consider a circular ds DNA E.coli chromosome. Consider markers located at distances x1, x2 and x3 respectively from oriT.



What are we distributing? Good question. We are distributing 'breaks' to x-length chromosome segment populations.






In case of x1-lengthers,
P(0) = Probability of having 0-break x1-lengthers = e^-mx1
P(1) = Probability of having 1-break x1-lengthers = mx1 * e^-mx1
P(2) = Probability of having 2-break x1-lengthers = (mx1^2 * e^-mx1) / 2
P(3) = Probability of having 3-break x1-lengthers = (mx1^3 * e^-mx1) / 6
.
.
.

Now comes the good part..
We are interested in whether a marker within distance x is transferred. It will be transferred if there are zero breaks in the length x. Hence we are interested in the fraction of those x-lengthers which have zero breaks. We are interested in the zero term P(0).
P(0) = e^-mx1


Similarly, fraction of markers at x2 transferred to the recepient,
P(0) = e^-mx2 and
fraction of markers at x3 transferred to the recepient,
P(0) = e^-mx3.

If we plot the values of P(0) for x1-lengthers, x2-lengthers and x3-lengthers versus the distance x1, x2, x3 then obvioulsy an exponentially decaying curve would be obtained. Multiplying the P(0) values of each by the total colonies scored will give the expected number of colonies recombinant for these markers, decaying exponentially with each's distance from oriT.