Premium
A Statistical Method for Comparison of Two Structures and Its Biological Applications
Author(s) -
Rachman M. I.,
Kozyakov S. Ja.
Publication year - 1986
Publication title -
biometrical journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.108
H-Index - 63
eISSN - 1521-4036
pISSN - 0323-3847
DOI - 10.1002/bimj.4710280208
Subject(s) - object (grammar) , homogeneous , computer science , incidence matrix , matrix (chemical analysis) , sampling (signal processing) , mathematics , algorithm , data mining , statistics , theoretical computer science , artificial intelligence , combinatorics , engineering , structural engineering , computer vision , filter (signal processing) , node (physics) , materials science , composite material
Some experimental methods applied to biological investigations allow to find a number q of homogeneous discrete components that determine a structure of the object in the study and the to establish the presence of some components from these q in compositions of other n objects. Such data form a nXq incidence matrix. In this paper a statistical approach is suggested for revealing the associated components in two different structures of the object. The procedure is based on joint consideration of two sampling incidence matrices. An application of this method is demonstrated by analysis of relationships between antigens and proteins of one strain of cyanobacteria. The method may be useful in different scientific areas, e.g. psychology, when direct experimental comparison of components from two structures is difficult.