Quantitative assessment of relationship between sequence similarity and function similarity

Abstract

Background: Comparative sequence analysis is considered as the first step towards annotating new proteins in genome annotation. However, sequence comparison may lead to creation and propagation of function assignment errors. Thus, it is important to perform a thorough analysis for the quality of sequence-based function assignment using large-scale data in a systematic way.

Results: We present an analysis of the relationship between sequence similarity and function similarity for the proteins in four model organisms, i.e., Arabidopsis thaliana, Saccharomyces cerevisiae, Caenorrhabditis elegans, and Drosophila melanogaster. Using a measure of functional similarity based on the three categories of Gene Ontology (GO) classifications (biological process, molecular function, and cellular component), we quantified the correlation between functional similarity and sequence similarity measured by sequence identity or statistical significance of the alignment and compared such a correlation against randomly chosen protein pairs.

Conclusion: Various sequence-function relationships were identified from BLAST versus PSI-BLAST, sequence identity versus Expectation Value, GO indices versus semantic similarity approaches, and within genome versus between genome comparisons, for the three GO categories. Our study provides a benchmark to estimate the confidence in assignment of functions purely based on sequence similarity.

Figure 1

Distribution of yeast and Arabidopsis unique orthologous pairs from COGs against sequence identity and expectation value intervals.

Figure 2

Relation between functional similarity in terms of the GO Indices and the negative logarithmic (base 10) E-value of sequence similarity within the same genomes using FASTA for the GO Biological Process Annotations.

Figure 9

Relation between functional similarity for GO Biological Process, Molecular Function and Cellular Component Annotations vs. E-value intervals (negative logarithmic with base 10) within the same genomes using PSI-BLAST.

Figure 3

Relation between percentage of sequence similarity and functional similarity for GO Biological Process Annotations within the same genomes using BLAST.

Figure 4

Relation between percentage of sequence similarity and functional similarity for GO Molecular Function Annotations within the same genomes using BLAST.

Figure 5

Relation between percentage of sequence similarity and functional similarity for GO Cellular Component Annotations within the same genomes using BLAST.

Figure 6

Functional conservation patterns for GO Biological Process annotations (A) based on evidences from experimental validations and (B) based on computational techniques such as electronic annotations, against percentage of sequence similarity.

Figure 7

A. Relation between E-value intervals (negative logarithmic with base 10) of seq uence similarity and similarity in SubLoc predicted localization of proteins within the same genomes using FASTA. B. Relation between percentage of sequence similarity and similarity in SubLoc predicted localization of proteins within the same genomes using BLAST.

Figure 8

Relation between Semantic similarity and sequence identity for GO Annotations for combined inter and intra genome comparisons using BLAST.

Figure 10

Relation between percentage of sequence similarity and functional similarity for GO (A) Biological Process, (B) Molecular Function and (C) Cellular Component Annotations within the same genomes using BLAST and (D) for all annotations using PSI-BLAST respectively, in the form of normalized ratio of pms(t1, t2), which is the probability of the minimum subsumer for terms t1 and t2 (see section 4.3).

Figure 11

Relation between percentage of sequence similarity and functional similarity for the GO Biological Process Annotations for inter-genome comparison of yeast ORFs against others using BLAST, in the form of normalized ratio.

Figure 12

Relation between percentage of sequence similarity and functional similarity for the GO Molecular Function Annotations for inter-genome comparison of yeast ORFs against others using BLAST, in the form of normalized ratio. Data points with a sample size less than 10 gene pairs are not sure, as the statistics is not significant.

Figure 13

GO Biological Process sub-graph with probabilities and minimum subsumer. The numbers in parentheses denote the occurrence of the GO term and any of its descendants in the GO.