A probabilistic method for identifying start codons in bacterial genomes

Abstract

As the pace of genome sequencing has accelerated, the need for highly accurate gene prediction systems has grown. Computational systems for identifying genes in prokaryotic genomes have sensitivities of 98-99% or higher (Delcher et al., Nucleic Acids Res., 27, 4636-4641, 1999). These accuracy figures are calculated by comparing the locations of verified stop codons to the predictions. Determining the accuracy of start codon prediction is more problematic, however, due to the relatively small number of start sites that have been confirmed by independent, non-computational methods. Nonetheless, the accuracy of gene finders at predicting the exact gene boundaries at both the 5' and 3' ends of genes is of critical importance for microbial genome annotation, especially in light of the important signaling information that is sometimes found on the 5' end of a protein coding region. In this paper we propose a probabilistic method to improve the accuracy of gene identification systems at finding precise translation start sites. The new system, RBSfinder, is tested on a validated set of genes from Escherichia coli, for which it improves the accuracy of start site locations predicted by computational gene finding systems from the range 67-77% to 90% correct.

Coding theory based models for protein translation initiation in prokaryotic organisms

“…The success of this work can lead to the development of improved methods for identifying the precise location of translation initiation start sites, an area that is receiving greater computational research attention (Frishman et al, 1999;Tompa, 1999;Hannenhalli et al, 1999;Suzek et al, 2001;Walker et al, 2002;Zien et al, 2000;Yada et al, 2001;Besemer et al, 2001). Additionally, design of effective coding-based models for genetic regulatory systems can potentially help researchers determine how to incorporate deliberate, sequence-controlled regulation into engineered proteins.…”

Correlation of mRNA Expression and Protein Abundance Affected by Multiple Sequence Features Related to Translational Efficiency in Desulfovibrio vulgaris: A Quantitative Analysis

“…To determine the effects of SD sequence during protein translation, we calculated the free energy for base pairing of 16S rRNA with SD sequence for each gene. (Suzek et al 2001): This method used a ''seed'' sequence to train a probabilistic model of SD sequences, which was then used to find the SD sequences in regions upstream of start codons of all genes. A good seed sequence is the 39 end of the 16S rRNA (Suzek et al 2001).…”

“…(Suzek et al 2001): This method used a ''seed'' sequence to train a probabilistic model of SD sequences, which was then used to find the SD sequences in regions upstream of start codons of all genes. A good seed sequence is the 39 end of the 16S rRNA (Suzek et al 2001). Five copies of 16S rRNA genes have an identical 39 tail in D. vulgaris (notably the annotated sequences were incomplete and missing the sequence ctggatcacctccttt at the 39 end; however, this sequence does exist in all five copies of D. vulgaris 16S rRNA genes).…”

“…The ribosomal binding to this purine-rich SD region is of prime importance to locate the ribosome at the proper initiation codon (Stenstrom et al 2001). The strength of this interaction has been used to estimate the efficiency of translation initiation (Schurr et al 1993;Osada et al 1999) or to predict the actual start sites (Suzek et al 2001). Second, nonrandom use of synonymous codons in the coding region of highly expressed Escherichia coli genes indicates that sequences further downstream of the start codon could be of importance for translation efficiency (Faxen et al 1991;McCarthy and Brimacombe 1994;Collins et al 1995).…”

A novel intracellular mutualistic bacterium in the invasive antCardiocondyla obscurior

“…The genome was searched against Rfam v.11 (Burge et al, 2013) using Infernal v.1.1 (Nawrocki and Eddy, 2013), and the resulting ncRNAs were manually integrated into the annotation following the INSDC conventions (http://www.insdc.org/files/feature_table.html). Ribosome-binding sites were predicted using RBSfinder (Suzek et al, 2001) and signal peptides were predicted using SignalP v.4.1 (Petersen et al, 2011). The resulting annotation was manually curated in Artemis (Rutherford et al, 2000).…”