Sequence pattern recognition in genome analysis

被引:0
作者
Luo, Liaofu [1 ]
Lu, Jun [1 ]
机构
[1] Inner Mongolia Univ, Fac Sci & Technol, Lab Theoret Biophys, Hohhot 010021, Peoples R China
来源
COMPUTATION IN MODERN SCIENCE AND ENGINEERING VOL 2, PTS A AND B | 2007年 / 2卷
关键词
pattern recognition; information extraction; information integration; IDQD; IDUD;
D O I
暂无
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
The problem of pattern recognition in genome analysis is studied. How the sequence information is extracted and integrated in the approach to sequence pattern recognition is discussed in detail. We propose two methods for calculation and prediction. The first is the Information Deviation Measure with Quadratic Discriminant (IDQD) and the second is the Information Deviation Measure with U-transformation Discriminant (IDUD). The former is applicable in case of sequence information obeying Gaussian-type distribution and the latter can be used in more general statistical distributions of sequence information.
引用
收藏
页码:1278 / 1281
页数:4
相关论文
共 8 条
[1]  
JIA Y, IN PRESS INNER MONGO
[2]   ON INFORMATION AND SUFFICIENCY [J].
KULLBACK, S ;
LEIBLER, RA .
ANNALS OF MATHEMATICAL STATISTICS, 1951, 22 (01) :79-86
[3]   MEASURE OF DIVERSITY [J].
LAXTON, RR .
JOURNAL OF THEORETICAL BIOLOGY, 1978, 70 (01) :51-67
[4]   Predicting conotoxin superfamily and family by using pseudo amino acid composition and modified Mahalanobis discriminant [J].
Lin, Hao ;
Li, Qian-Zhong .
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS, 2007, 354 (02) :548-551
[5]  
Lü J, 2005, PROG BIOCHEM BIOPHYS, V32, P1185
[6]  
LU J, 2006, P CHIN ASS SCI TECHN, P791
[7]   Splice site prediction with quadratic discriminant analysis using diversity measure [J].
Zhang, LR ;
Luo, LF .
NUCLEIC ACIDS RESEARCH, 2003, 31 (21) :6214-6220
[8]  
ZHANG YJ, IN PRESS ACTA SCIENT