Archive for August, 2007

Really, what is gene?

Monday, August 27th, 2007

After years of studying and doing research on biology, I sometimes (always?) feel frustrated when I tried to answer one question as simple as it might sound, what is gene?

Frankly speaking, I didn’t reach that frustration until I read one paper/book by Francis Crick. He was arguing something else with the example of how hard it would be even if presumably biologists are trying to just explain a very simple thing/concept–what is gene? I didn’t fully buy it at very beginning, especially, I considered myself have worked on biology for quite a while. I tried. At first, I got a very simple answer, then, I had to modify myself due to the fact that in biology, whenever people establish a rule, there will always be exception, bit by bit, day by day, finally, I ended up with a awkwardly tedious answer and questioning myself, do I know biology at all?

  • Back to 1900s, Johansson introduced gene as abstract explanation for the hereditary basis of traits.
  • George Beadle (he graduated from where I’m :-( ) claimed “one gene, one enzyme” -> “one gene, one peptide”.
  • The emergence of RNA caused troubles.
  • There is a saying from an American president(not sure which one):

    If you can’t solve the problem, magnify it!

    A gene is a DNA sequence segment which is responsible for making a functional unit (synthesize a protein product or regulatory product like RNA).

    Can we identify all of them in genomes, for example, human genome?

    Excellent videos on machine learning

    Thursday, August 23rd, 2007

    Just realized that there are a number of excellent sites offering very up-to-date hot spots on machine learning, or, knowledge discovery. Well, they are all for those who missed the conferences. (I missed all of them. :-( )

    KDD2007
    ICML 2007

    Personal medicine continue

    Wednesday, August 22nd, 2007

    This topic has been heat up. I guess because we all are looking for next exciting spot in biology/medicine, especially from industry perspective. Here is one article on the topic again.

    VentureBeat

    Navigenics, a new personal genetics startup with some serious backing, threw back the curtain over the weekend by unveiling its Web site. The Redwood Shores, Calif., startup says it aims to provide individuals with their genetic profiles and then to “arm” them with ways to improve their future heath.

    This is very similar to what 23andMe, a similar startup backed by Google and Genentech (see our coverage here and here), intends to do. Adding to the intrigue is the fact that Navigenics has some influential supporters of its own, including Kleiner Perkins Caufield & Byers and Sequoia Capital. KP’s uber-VC John Doerr also has a seat on its board. KP, of course, was an early backer of Google, whose co-founder Sergey Brin just happens to be married to 23andMe co-founder Anne Wojcicki.

    Even it is still emerging, we already see competitors and “gangs” of competitors.

    Navigenics + Affymetrix VS 23andme + Illumina

    If we explicitly put Google and Genentech on the side of 23andme, it would look too scary. Right?

    Dr. Personal Medicine

    Tuesday, August 14th, 2007

    People, at least within the biomedicine community has been talking about personal medicine for years if not decades. The actual delivery is way over due. It seems wherever we were left by the slowing moving research and policies regulating the process and implementation is going to be picked up by two giants in industry, Google and Microsoft. For example, on today’s New York Times, there is an article titled as “Google and Microsoft Look to Change Health Care” discussing the new trend in information and software companies. Here is the quotation from the article.

    In politics, every serious candidate for the White House has a health care plan. So too in business, where the two leading candidates for Web supremacy, Google and Microsoft, are working up their plans to improve the nation’s health care.

    This is a great news for people who work in biomedicine, at least, though the customs/patients will undoubtedly benefit the most. I have to admit that health information. For example, we always have problems with our own medical records, they might be lost, and hardly comprehensive, very localized, that is, whatever diagnosis and treatment we get will not be easily transfered to other hospitals, background and expertise of doctors and hospitals, and the very messy insurance companies. Another thing is that we, at least myself always search for healthy issues, such as, how to lose belly fat? The best answer will have to be somehow personalized which is not possible from current search engines. It is therefore what computers can help us out and what we, bioinformatics can fit for such a demand.

    Scifoo dream

    Sunday, August 12th, 2007

    The scifoo was over without my attending. :-( After read many posts by those who attended the unconference, I can feel the warm messages from the data analysis community. Especially, the inspiring experience from other established persons and startups, the aggressive and somehow ambiguous plans/pictures which are just emerging, are very impressive and promising. After decades collection, it seems every industry have been collecting massive amount of data and need to be analyzed in order to understand and predict either the commercial, social, or natural patterns underneath the mountains of data. Good to know.

    One of my dreams is to attend one scifoo in future.

    To continue the trend of getting fresh air from the community, I got a RSS sites for news in journals and science, which is supposed to be the up-to-date collection, check out:

    BioMed Central RSS Feeds

    I have to say that the bioinformatics world is changing in a surprised pace.

    An interesting event

    Wednesday, August 1st, 2007

    I just realized that there is an event that I don’t want to miss, but, I’ll miss it for sure. It is the second scifoo camp, which is organised by Nature Publishing Group, O’Reilly and Google. It will be held at the Googleplex headquarters in California from Aug.3 to Aug.5, 2007. Since this is an invitation type of event and I didn’t receive the invitation yet, :-( , I’m gonna to miss it for sure. D’uh…!