Archive for October, 2007

References for machine learning

Monday, October 29th, 2007

It is a good way to record references which might be inspiring some other days. Here is my current collection of good references for machine learning:

A tutuorial on hidden Markov models and selected applications in speech recognition
Lawrence R Rabiner

Introduction to statistical learning theory
Olivier Bousquet, Stephance Boucheron, and Gabor Lugosi

Neural networks and machine learning in bioinformatics-theory and applications
Udo Seiffert, Barbara Hammer, Samuel Kaski, and Thomas Villmann

Statistical mechanics of learning from examples
H.S.Seung, H. Sompolinsky, and N. Tishby Physical Review A

From data mining to knowledge discovery in databases
Usama Fayyad, Gregory Piatetsky-Shapiro, and Padhraic Smyth

Knowledge Discovery in Databases: an Overview
William J. Frawley, Gregory Piatetsky-Shapiro, Christopher J. Matheus

Machine learning in bioinformatics
P. Larranaga, …, V. Robles

Systems biology reviews

Saturday, October 27th, 2007

There were reviews regarding the systems biology in Nature 2006 and I found they are still quite interesting.

A number of books have been published on systems biology. I kinda agree with that systems biology makes more sense than just collection individual genes and proteins. One book that I am reading was written by Bernhard O. Palsson, which titles “Systems biology: properties of reconstructed networks“. There is also another one by Uri Alon: An Introduction to Systems Biology: Design Principles of Biological Circuits (C&H/CRC Mathematical & Computational Biology Series).

Some papers that I read also deserve to recommendation:

Structural systems biology: modelling protein interactions
Patrick Aloy and Robert B. Russell Nature 2006

Structural bioinformatics
Rob Russell and also this.

Network motifs in the transcriptional regulation network of Escherichia coli
Shai S. Shen-Orr, …, Uri Alon

Network biology: understanding the cell’s functional organization
Albert-Laszlo Barabasi and Zoltan N. Oltvai Nature reviews 2004

Computational prediction of eukaryotic protein-coding genes
Michael Q. Zhang Nature 2002

Seven-transmembrane receptors
Kristen L. Pierce, …, Robert J. Lefkositz Nature reviews 2002

One example of the application to microbial kinome:
Structural and funtional diversity of the microbial kinome
Natarajan Kannan, …, Gerard Manning PLOS Biology 2007

One of my questions to systems structural biology is how to interpret pathways in terms of networks of structures?

Green computation

Wednesday, October 24th, 2007

Nowadays, whichever university I go, there are always clusters of supercomputers. Well, maybe because I always go to those places with a stress on computational applications. But overall, that tells me that we have many of those energy costing machines. For examples, those machines actually cost a lot to run and at same time cool themselves. They need to constantly generate useless heat and not burn themselves. The heat product has nothing to do with the computation which is what those machine meant to do. The need for more supercomputers is still accumulating at a fast pace for weather prediction, drug discovery, various simulation used by the government, industrial labs,… Given these concerns, I’d like to propose next generation computation in my cartoon.

Cao’s next generation computation

Environment friendly
Energy efficient
Operation remain same (remote access)
Not abandon well adapted parts (Ethernet …)
Decorating the sky (in a nice way)

High maintainance fee (especially at beginning)
Can’t touch them anytime you want

Conclusion: Good to go! :-(

The forgotten code cracker

Monday, October 22nd, 2007

I’m not a historian, at least, a biology historian. For example, I really didn’t think Marshall W. Nirenberg deciphered the genetic code. Quite frankly, I thought Francis Crick and/or James Watson did so. I just got another side of the story from the latest Scientific American (November 2007). The article is titled “The forgotten code cracker”, it claims that Robert W. Holley, Har Goind Khorana, and Marshall W. Nirenberg were awarded the Nobel Prize in Physiology or Medicine in 1968 “for their interpretation of the genetic code and its function in protein synthesis”.

Besides the vivid description of the history of deciphering the genetic code, I’m also impressed with the last sentence in the article:

Fame may be fleeting, but the genetic code will endure for as long as there is life.

The author is Ed Regis, his new book is “What is life?(Farrar, Straus and Giroux 2008)”.

Disordered protein

Saturday, October 20th, 2007

There is a very good paper on disordered protein, which I think it is quite interesting.

Intrinsically disordered protein

A computational method to predict disordered proteins is here:

References to graph theory and biology

Thursday, October 18th, 2007

I have thought about learning some graph theory for a long time. Finally, I was able to do some searches for references. luckily, I got some and would like to share them here:

Understanding network concepts in modules
Jun Dong and Steve Horvath

Complex Networks: from Graph Theory to Biology
Graph Theory and Networks in Biology
Oliver Mason and Mark Verwoerd

Evolutionary graph theory: Breaking the symmetry between interaction and replacement
Graph theory and networks in biology
Using graph theory to describe and model chromosome aberrations
Graph theory and biology

Although I heard that many people are very interested in the application of graph theory in biology, I’m still skeptical about it. The promising of the application arise from the fact that biological phenomena is resulted in the complicated networks of molecular activities. The trouble is that, IMHO, we don’t know all the players in most of the networks. I’d like to wait for more experiments to think how to approach biology questions from this mathematically very nice graph theory.

Links: how to give an academia presentation

Tuesday, October 9th, 2007

Living in the digital world is really wonderful, you have tremendous access to information and resources and also there are many people out there doing very good and competitive jobs that could benefit what you are doing provide you are willing to communicate. Here are some useful links.

How to give a not bad academia presentation? [link]
A new conference on aging: [link]

Sodium ion channel

Saturday, October 6th, 2007

It is the first time that we have a solved 3-dimensional atomic structure for the sodium ion channel family. The work was published on Nature. The corresponding PDB deposit is 2QTS.

More, more structures for membrane proteins, please.

The debate on consciousness

Monday, October 1st, 2007

In the latest issue of Scientific American(Oct.2007), there is a fascinated debate on how to determine the neuronal correlates of consciousness, in other words, how do we map specific conscious experiences to brain activity. I’d like to brief their arguments and reasoning here.

There are numerous phenomena in consciousness need to be explained.

  • Self consciousness which examine one’s own desires and thoughts
  • Content of consciousness which is that you are actually conscious of at any moment
  • Relation of brain process to consciousness and to nonconsciousness
  • To make this field even more appealing to people who want to step into neuroscience, we don’t know how the brain works to implement consciousness arising from the electrical and chemical activity of neurons. The unknown list could be much longer than here. But that is not what the debate is focus. They are actually what the two researchers agree on. The center of this debate is related to what happens after the consciousness arises — to determine the best neuronal correlates of consciousness. They are trying to explain, each in his/her own ways how the brain matches up with specific conscious experiences.

    Christof Koch vs Susan Greenfield

    What is the activity in the brain during subjective experience?

    The brain has specific groups of neurons mediating, or even generating, distinct conscious experiences. An argument is that organisms evolve specific gadgets, and it is true for the brain.

    “A specific network of neurons is needed for a specific percept, not any random collection of neurons that become highly active. Furthermore, for full consciousness, a coalition of neurons must encompass both sensory representation at the back of the cortex as well as frontal structures involved in memory, planing and language.”

    “Consciousness is generated by a quantitative increase in the holistic functioning of the brain.”

    Before converted to biology, I would have agreed with Greenfield in a substantial degree. However, after reading numerous books and papers, listening to some big shots’ lectures/conversations, and seeing the beauty of biology, chemistry, and physics, I appreciate Koch’s arguments a lot. In the meanwhile, I have to say that I’m not totally convinced by his arguments and opinions. I’ll post my uncomfortable thinkings some other time.