Saturday, March 31, 2012

03/31/12

这个周末开会,mike 70岁生日和simon60岁生日加上MCB项目10周年的聚会。牛人来了不少,David Haussler,Wing H Wong, Terry Speed, Barry, Ron 好多牛人,还有很多MCB毕业的学生博士后。看着自己桃李满园,他俩一定特开心,虽然不是所有的talk质量都很好。

遇到了一个越南人,在圣彼德堡大学和UCSD作博士生,看起来很学术,他是N同学的硕士毕业时候的指导者之一。世界还挺小的。他的导师是个很有意思的俄国人,二十多年前来USC跟mike和simon讲De bruin graph,当时苏联还不允许外国货币进入苏联,他就只带了15马克在洛杉矶待了一周,到了LA,mike接他到宾馆,Pavel很兴奋,跟mike说我们来讨论bioinformatics吧,mike很diplomatic地跟他说“你累了”,他很憋屈,凌晨4点睡不着就到Downtown去逛了一圈。第二天,m和s知道了这事儿之后就明白"they're in trouble"了。 等P同学给了talk之后,m和s明白了几件事,P knows nothing about possion approximation, and he doesn't speak english.


今天晚上去K和J家吃饭,I同学带我去的,玩guitar hero,draw sth。我觉得K对I格外好。

Monday, February 27, 2012

ggplot

> head(tmp)
    V1     V2     V3 hcg     hmeth mch      mmeth class       h_m
1 chr1 884000 885000 107 0.0105356  55 0.00406504     M 0.2890211
2 chr1 885000 886000 103 0.0158201  30 0.01015230     M 0.5357160
3 chr1 886000 887000 119 0.0170750  36 0.41832700     L 0.5192445
4 chr1 887000 888000  50 0.5725810  16 0.94303800     L 0.4948500
5 chr1 888000 889000  51 0.8819190  29 0.95502600     M 0.2451722
6 chr1 889000 890000  73 0.8581730  21 0.94063900     M 0.5411036






pdf("Human_mouse_cg.pdf")
ggplot(tmp, aes(x=h_m, colour=class)) + geom_density(adjust=5)
dev.off()


where the adjust  option adjusts the smoothness of the density plot.




http://had.co.nz/ggplot2/stat_density.html






Change axis labels 

By default, ggplot uses the variable name as the axis labels. Change it to something else using scale_x_continuous or scale_y_continuous
p = ggplot(diamonds)+geom_point()+aes(x=carat,y=price)

p + scale_x_continuous('x axis label') + scale_y_continuous('y axis label')


Colors
look at here:
http://wiki.stdout.org/rcookbook/Graphs/Colors%20(ggplot2)/

Wednesday, February 22, 2012

02/22/12

好吧,最近不爽,很不爽。科研停滞不前,跟老板沟通有障碍。我觉得我就是爱走极端,陷到自己的情绪里无法自拔。真傻,怎么能跟老板怄气呢,老板对我那么好,怎么能把对自己的不满发泄到老板身上呢? 好想哭,乱七八糟的,越叫我有条理我就越乱。

今天09年化学诺奖得主跟大家聊天,嗯 好奇心最重要,可是我现在的好奇心降到奇低的水平,没有方向,何言进度!怎么能洗心革面,重新做研究僧呢? 想骂街骂街骂街骂街!

Wednesday, February 1, 2012

01/02/12

在冷泉实验室,昨天一切顺利,跟着安团做了两三步。今天BStreatment也还好,可是到下午最关键时候,安团同学去开会了剩下我跟艾米莉做Egel,然后我就倒霉了。

明天给大家讲HMM,希望不会掉链子。

Wednesday, January 11, 2012

01/11/12

1. 要解决的问题
2. 现行方法缺陷
3. 创新之处,为什么新方法能解决问题
4. 新方法的应用前景

Wednesday, January 4, 2012

Paul Baran

The chances are you've never heard of him, but without his pioneering work in the 1960s, you wouldn't be reading this web page, because there wouldn't be a web for you to read it on.
His name is Paul Baran, and he died March 26, aged 84. Baran, a Polish-born engineer at the RAND corporation, invented one of the most fundamental aspects of internet technology, the concept that later became known as "packet switching".
Put simply, packet switching is the way data is chopped up into smaller chunks, and each chunk is sent across a distributed network to end up at its destination.
The chunks, or packets, might not all take the same route. The whole point is that there are many different routes available to them. Each packet has something called a header — a summary of metadata, information about the packet and its contents.
As the packets arrive at their destination, the packet headers tell the computer there how to re-assemble them in the right order.
Baran shares some of the credit for packet switching with researchers Donald Davies and Leonard Kleinrock, who were working independently along similar lines; before they came up with the idea, information had to go from point A to point B using what's called circuit switching. (Think of old footage of manual telephone exchanges, where operators plugged wires to connect one circuit with another.) With packet switching, a single circuit could be used more efficiently, to send more data between more people.
Baran's pioneering work in this field was a direct precursor to the creation of the government-sponsored network Arpanet, which itself was one of the first building blocks of the internet as we know it today.
A version of this text originally appeared on TIME.com on March 28, 2011.


Read more: http://www.time.com/time/specials/packages/article/0,28804,2101745_2102136_2102369,00.html #ixzz1iY27cuxm

Monday, January 2, 2012

纽约Day14

起来以后到星巴克吃了早餐,然后去Path站门口的UPS店打印了帝国大厦的票和rent的票,打印要50美分一张,真是贵,唉。然后直奔中央公园,在湖边转了小半圈以后做公车到39街的草堂小餐吃午饭,点了干煸四季豆,干烧鱼片,和春笋排骨汤。午饭后已经三点左右了,走路到42街看了一下中央车站 Grand Central Station,昊爷在那儿花了不到2分钟时间买了一个iPad。然后继续走到33街的帝国大厦,排队过安检,再排队上电梯,再排队爬楼梯到86层,看了落日,然后看了夜景,又走回草堂小餐要了碗酸辣粉,还点了钵钵鸡和肉末酸豆角,然后奔去50街看Rent,中间还在一家店里买了一小碗冰淇凌。Off-broadway的阵容果然不够强大,Mimi除了出场的一首歌唱的还行,其他的solo嗓音都不亮,有点失望。不过Maureen很给力,又能演又能唱,Angel也是,竟然是个高中生,不知道是不是真的gay。