Database Reference
In-Depth Information
TABLE 8.2
Types of biological data and analysis techniques used for
predicting protein function
Data Type
Analysis Techniques
Amino acid
sequences
Flexible sequence similarity measures and clustering
Classification based on subsequences such as motifs and
domains
Classification based on biological features derived from
sequences
Protein
structure
Structural similarity-based inference
Inference using 3D structural motifs
Inference using features of the 3D surface of a protein
Classification using structure kernels and frequent
substructures
Genome
sequences
Proximity of two genes and their orthologs in multiple
genomes
Fusion of two proteins into a single protein in other
genomes
Evolutionary
data
Co-occurrence of two genes in multiple genomes
Functional inference based on duplication and speciation
events
Microarray
data
Clustering for finding functional groups
Classification to infer functions of individual proteins
Classification of temporal microarray data
Protein
interaction
networks
Annotation transfer from neighboring proteins in
network
Global annotation transfer from the whole network
Clustering to find densely connected regions
Association analysis to find frequently occurring
subnetworks
Biomedical
literature
Information retrieval using word frequency-based
statistics
Text mining using classification and clustering of
documents
Natural language processing-based approaches
Multiple data
types
Combination of multiple data types into a single graph
Combination of predictions from different data types
Intelligent fusion of multiple datasets of different types
