发信人: Maxiwell (东方乙), 信区: Biology
标 题: Re: [转载] Re: Bioinformatics 冬天已经来了--它的春天在哪里?
发信站: The unknown SPACE (Thu May 9 22:17:18 2002) WWW-POST



早在两年前的时候 nature biotechnology 就发出过对这个领域的前景报导的一
篇文章, 当时也是一个小的bioinformatics公司关门了. 记得还是个女abc当老板什么
的. Anyway, 撇开别的不说, 关键在于business model的问题, 很多公司想用给
licence的方法来拉住大 至药厂客户, 但业内人士对于在internet上传送公司机密
sequence的idea 一向不屑一故. 放开security不说, 这种只对big pharm的市场实在
是太小了. 有点类似 Oracle的 方式, 但oracle的用户市场开阔的多, 但是就licence
的model 来说, Oracle的日子也不是很好过.

Anyway, 从整个pharm的角度来讲, 其实它的市场也只是病人, focus在保健, 和疾病
上, 比起广泛的computer 的市场来说, 本身就小了一些. 要成功, 一定要看好有大的市
场
另外一定要领先对手, 积极创新. viagra其实只针对男人, 本身市场就砍掉一半, 女人
不适用, 它的成功在于它是首例, 领先了对手. 扯远了, 再谈bioinformatics. 它的
市场不大, 而且不容易在竞争中有自己的拳头产品, 有的只是hype. 所以是一个泡泡.


不过bioinformatics的确是一个有前途的方向. 我认为在 data 处理的方向上
仍然大有可为. 这一点将很接近现在的生统, 生统其实focus在临床药性 BMPK方面.
如果bioinformatics能将 临床和基础的数据 很好的分析出来, 是非常有用的.
设想一下, 数年内如果 实验技术发展到每人每天可作5000个point mutation, 一万
次assay, 十万个 gene cloning. 那么一个礼拜之内, 将出现 tons of data.
人为分析这些数据, 将是不可能的. 所以必须要bioinformatics.


但是目前实验, 还是labor intensive, 不可能上规模. 而且实验结果的quality 没有
safe guard. 此时用bioinformatics 首先不满足统计学上sample数量的要求, 另外
data不可信, 对于CS也好, 更无从说起, 这就叫做 garbage in garbage out.
不管你bioinformatics tool 有多好.

bioinformatics说到底, 也是门science 而不是 engineering, 或者 business
大家一窝疯上去炒, 嘿嘿. 难道不知道scientist 通常是穷光蛋的道理么?


anyway, 胡说八道几句. 算是马后炮吧. 但愿不准.
















【 在 leohawk (leohawk) 的大作中提到: 】
: yeah, agree, but I am sure what optimist tried to express is not
: what it seems to be.
:
: I think most Math and CS people is trying to building something up very
: systematically, mostly by generalizing something out from simple facts...
: this is typical Western style of doing science, and probably the only good
: way to do it. It is slow, but it is an acceleration process.
:
: It will take long to develop theory out that applies to all bio phenomena,
: but once there is such thing ( I can't see it coming though ), it is
: going to be good. This will requires way extended understand of biology
: using experimental methods.
:
: biological research is still a case by case kind of research, one gene one
: postdoc/graduate student pattern won't change much in the near future,
: otherwise, we won't be able to collect enough information out to apply
: a more systematic way of research, not to say applying MATH or CS to it.
:
:
: 【 在 accura (指导, 回归) 的大作中提到: 】
: : hhe, although I agree that we need a deep understanding of biology,
: : I donot agree ur extend idea. :)
: : I donot think they make simple things complicated, rather, they tried to
: : develop tool from simple thing.
: : 【 在 optimist (乐观派) 的大作中提到: 】
: : : Exactly. I can't agree more.
: : : Regarding your last statement,that's what I feel too.
: : : hehe...mathematician-bioinformatist just like to put all bunch of
formulars on
: : : their paper or books which only make actual simple things
complicated.Although
: : : I doubt their real use to biology,but people expect their contribution
because
: : : nobody understand what they are talking about...sorry,no
offense,but...hehe
: : : ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
: : : bioinformatics
: : : many
: : : once
: : : them
: : : bioinformatics,
: : : bioinformatics to
:
:


--
※ 来源:．The unknown SPACE bbs.mit.edu．[FROM: 63.208.131.238]