个交叉学科的胜利■ 简介 生物信息学 08300240054黄晓靖
一个交叉学科的胜利 简介 生物信息学 08300240054 黄晓靖
简介生物信息学 现代医学的 生物信息学 分子同源性 挑战 计算 一个交叉学科的胜利
简介 生物信息学 一个交叉学科的胜利 现代医学的 挑战 生物信息学 分子同源性 计算
现代医学的挑战 一个交叉学科的胜利
现代医学的挑战 一个交叉学科的胜利
现代医学的挑战 过去 人 经验学科 ·现在 Global circulation of influenza viruses mber of specimens positives for influenza by subtypes ·极小 week 17 (2009)-19 (20to)from 19 April 2009 to 15 May 2010 极大 ·实验学科 tHlluumillnlHHwwinmapRrfil 2品品品E2222
• 过去 • 个人 • 经验学科 • 现在 • 极小 • 极大 • 实验学科 现代医学的挑战
现代医学的挑战 大规模数据的采集 大规模数据的处理 经验型处理常常会出错 Wa/-Mart About WAL*MAR
• 大规模数据的采集 • 大规模数据的处理 • 经验型处理常常会出错 • Wal-Mart 现代医学的挑战
现代医学的挑战 cs to the rescue
• CS to the rescue~~~ 现代医学的挑战
生物信息学 一个交叉学科的胜利
生物信息学 一个交叉学科的胜利
Bioinformatics 当生物遇上了计算机 生物信息学!
当生物遇上了计算机 生物信息学！ Biology Informatics Bioinformatics
网络预测流感 ublished online 19 November 2008 Nature 456, 287-288(2008)I 0oi:10.1038/456287 Web data predict flu Search engines provide information about epidemics TRACKING THE FLU Declan Butler The relative frequency of flu-related key words in Google searches closely tracks flu statistics in the US mid-Atlantic region as monitored by government officials. Two new studies hint at the public-health and research potential of 12 mining the data created as people search the web. Both teams have Google flu trends uccessfully detected the onset of us seasonal fiu epidemics, by 8v Centers for Disease Control extracting pattems of flu-reiated search terms from the bilions of and Prevention data quenes stored by Google and Yahoo The work tested the hypothesis that people will more frequently search the Intemet using flu-related terms when they get sick, One group used Google's search-query logs, the other Yahoos. Together they generated strikingly concordant findings: patterns of searches 八类A人 matched almost perfectly with official flu surveillance data - and often weeks in advance of these 2004 2006 2007 2008 ction with researchers at the University of lowa in lowa City rvard University, manually selected key words for testing 'flu or influenza while eliminating confounding terms avian' or 'bird' The researchers compared the relative frequency of the search terms, between 2004 and 2008, with weekly national data on the standard surveillance indicators of flu-positive viral isolates and flu mortality rates. ' We found that we could explain weekly variation in seasonal influenza one to three weeks in advance of cultures, and five weeks in advance of mortality, says Philip Ofgreen epidemiologist at the University of towa
网络预测流感
nature Vol 457 19 February 2009 doi: 10. 1038/nature07634 LETTERS Detecting influenza epidemics using search engine query data Jeremy Ginsberg, Matthew H Mohebbi, Rajan S Patel, Lynnette Brammer, Mark S Smolinski& Larry Brilliant Seasonal influenza epidemics are a major public health concern, By aggregating historical logs of online web searchqueries submitted causing tens of millions of respiratory illnesses and 250,000 to between 2003 and 2008, we computed atime series of weekly counts for 500,000 deaths worldwide each year. In addition to seasonalinflu- 50 million of the most common search queries in the United States enza, a new strain of influenza virus against which no previous Separate aggregate weekly counts were kept for every query in eacl immunity exists and that demonstrates human-to-human trans- state. No information about the identity of any user was retained. Each mission could result in a pandemic with millions of fatalities: time series was normalized by dividing the count for each query in a Early detection of disease activity, when followed by a rapid particular week by the total number of online search queries submitted response, can reduce the impact of both seasonal and pandemic in that location during the week, resulting in a query fraction influenza. One way to improve early detection is to monitor (Supplementary Fig. 1) health-seeking behaviour in the form of queries to online search We sought to develop a simple model that estimates the probabil- engines, which are submitted by millions of users around the ity that a random physician visit in a particular region is related to an world each day. Here we present a method of analysing large ILl; this is equivalent to the percentage of ILl-related physician visits numbers of Google search queries to track influenza-like illness A single explanatory variable was used; the probability that a random in a population. Because the relative frequency of certain queries is search query submitted from the same region is ILI-related, as deter- highly correlated with the percentage of physician visits in which a mined by an automated method described below. We fit a linear patient presents with influenza-like symptoms, we can accurately model using the log-odds of an ILI physician visit and the log-odd estimate the current level of weekly influenza activity in each of an ILl-related search query: logit(l(0))=logit(Qn))+s where region of the United States, with a reporting lag of about one I(t)is the percentage of ILl physician visits, Q(n) is the ILl-related <y. This approach may make it pos sible to use search queries to query fraction at time t, a is the multiplicative coefficient,ande is the da etect influenza epidemics in areas with a large population of web error term. logit(p)is simply In(p/(1-P) arch
网络预测流感(2)