生物间相似性的多样性指标

  1. 1、下载文档前请自行甄别文档内容的完整性,平台不提供额外的编辑、内容补充、找答案等附加服务。
  2. 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
  3. 3、如文档侵犯您的权益,请联系客服反馈,我们会尽快为您处理(人工客服工作时间:9:00-18:30)。

数学建模网络挑战赛

承诺书

我们仔细阅读了第四届“互动出版杯”数学中国数学建模网络挑战赛的竞赛规则。

我们完全明白,在竞赛开始后参赛队员不能以任何方式(包括电话、电子邮件、网上咨询等)与队外的任何人(包括指导教师)研究、讨论与赛题有关的问题。

我们知道,抄袭别人的成果是违反竞赛规则的, 如果引用别人的成果或其他公开的资料(包括网上查到的资料),必须按照规定的参考文献的表述方式在正文引用处和参考文献中明确列出。

我们郑重承诺,严格遵守竞赛规则,以保证竞赛的公正、公平性。如有违反竞赛规则的行为,我们将受到严肃处理。

我们允许数学中国网站()公布论文,以供网友之间学习交流,数学中国网站以非商业目的的论文交流不需要提前取得我们的同意。

我们的参赛队号为:1007

参赛队员(签名) :

队员1:刘路

队员2:于涛

队员3:戴超逸

参赛队教练员(签名):

参赛队伍组别:本科组

数学建模网络挑战赛

编号专用页

参赛队伍的参赛队号:(请各个参赛队提前填写好):

1007(本科组)

竞赛统一编号(由竞赛组委会送至评委团前编号):竞赛评阅编号(由竞赛评委团评阅前进行编号):

2011年第四届“互动出版杯”数学中国

数学建模网络挑战赛

题目基于生物间相似性的多样性指标

关键词物种分类、熵、基因、多样性、相似性、

摘要:

目前的物种多样性测量方法绝大部分是依据物种丰富度或者平均度建立的。这些指标虽然能很好的反映物种种类总数和个体在各个物种间分布的均匀水平,但它们丢失了一个重要的信息,即没有考虑不同物种间相似性的程度,这将使生物多样性的本质含义不完整。本文针对这一缺陷给出两个模型,树模型和统计熵模型。

树模型中我们参照了生物学中的经典分类理论(按界、门、纲、目、科、属、种划分),给树形分类目录中不同等级的子节点定义一个权重,根据两种物种在分类树中对应的“路径”来计算每两种生物的差异程度;根据一种物种与其他物种差异程度的均值算出它与生物群落中的其他所有物种间的平均差异程度;最后对样本中所有物种的平均差异程度求和,作为该生物群落的多样性指标。文章中还验证了树模型具有计算简洁、易扩展(可推广)等良好的性质,例如还可以容易地将各物种所占比重也考虑进去。

统计熵模型借鉴了物理学中形容事物分布无序程度的熵概念,将生物的多样与其分布的无序类比,这一模型中体现了对生物间相似程度的比较。

对于两个模型我们都给出具体的检验数据并通过计算结果说明这两种指标在一定程度上优于传统及现存的测量方法,例如当某几种生物的比重占主导时,现存的测量方式使得新增物种对生物多样性的贡献甚微,而我们设定的新模型却在这种情况下对检测物种增减具有更强的灵敏性。

参赛队号 1007 所选题目 B 参赛密码

(由组委会填写)

Abstract

Presently, most of biodiversity indices rely on either richness or evenness. Although these indices more or less reflect the total number of species and the homogeneity of distribution of each species, but they all ignore one thing, i.e the degree of difference or similarity among species. This considerably drove these indices apart from the definition of biodiversity. In this article, we aim at this defect, and give two models, tree model and statistic entropy model.

In tree model, we refer to the classical classification theory ( i.e classify species by several levels including phylum, class, order, family, genus, species ), and we assign a weight to each node in the taxonomic tree. We calculate the difference degree between two species according to their corresponding path (and the weight of each node in this path) in the taxonomic tree. We define the difference between one specie and a set of species to be the average difference degree between this specie and each specie within the set. Finally, we define the biodiversity degree of this set of species to be the sum of the average difference (with respect to this set) of each specie in the set. We argued that tree model index has relative low complexity ( not as it seems to be ), high flexibility ( generalizable ) etc. For example, one could easily add to the model the weight of each species.

The statistic entropy model borrows idea from what is used to describe “disorder” of the distribution of a set of object in statistical physics, i.e entropy. We make analogy between biodiversity and the disorder of the distribution of a set of species. This model also reflects the degree of similarity among species.

We test our models and showed that these two models, to some extent, are more or less superior to the traditional and present measurements. For example, when several species dominates the whole set of species, present measurements ( i.e Shannon index, Simpson index ) varied fairly fractional, while tree model index does not suffer this.

相关文档
最新文档