TY  - BOOK
AU  - Fassetti,Fabio
AU  - Rombo,Simona E.
AU  - Serrao,Cristina
ED  - SpringerLink (Online service)
TI  - Discriminative Pattern Discovery on Biological Networks
T2  - SpringerBriefs in Computer Science,
SN  - 9783319634777
AV  - QH324.2-324.25 
U1  - 570.285 23
PY  - 2017///
CY  - Cham
PB  - Springer International Publishing, Imprint: Springer
KW  - Bioinformatics
KW  - Pattern recognition
KW  - Data mining
KW  - Gene expression
KW  - Computational Biology/Bioinformatics
KW  - Pattern Recognition
KW  - Data Mining and Knowledge Discovery
KW  - Gene Expression
N1  - Part I: Biological Networks -- Data Sources and Models -- Problems and Techniques -- Part II: Pattern Mining -- Exceptional Pattern Discovery -- Discriminating Graph Pattern Mining from Gene Expression Data; Available to subscribing member institutions only. Доступно лише організаціям членам підписки
N2  - This work provides a review of biological networks as a model for analysis, presenting and discussing a number of illuminating analyses. Biological networks are an effective model for providing insights about biological mechanisms. Networks with different characteristics are employed for representing different scenarios. This powerful model allows analysts to perform many kinds of analyses which can be mined to provide interesting information about underlying biological behaviors. The text also covers techniques for discovering exceptional patterns, such as a pattern accounting for local similarities and also collaborative effects involving interactions between multiple actors (for example genes). Among these exceptional patterns, of particular interest are discriminative patterns, namely those which are able to discriminate between two input populations (for example healthy/unhealthy samples). In addition, the work includes a discussion on the most recent proposal on discovering discriminative patterns, in which there is a labeled network for each sample, resulting in a database of networks representing a sample set. This enables the analyst to achieve a much finer analysis than with traditional techniques, which are only able to consider an aggregated network of each population
UR  - https://doi.org/10.1007/978-3-319-63477-7
ER  -