Prediction of probable genes by Fourier analysis of genomic sequences

Ab initio gene identification: Prokaryote genome annotation with GeneScan and GLIMMER

“…On the other hand, we find a large number (29) of False Positives. Three of these show a significant similarity to genes from other organisms such as Escherichia coli and Neisseria meningitidis, and the Fourier spectrum for 2 of them is "typical" of genes (Tiwari et al 1997). Sixteen of these appear to be possibly novel genes, while the remainder appear to be copies of other genes on the genome.…”

“…The three genomes used in this 3 Thus P N is further divided by a factor to make it independent of the choice of window length. This was not explicitly clarified in Tiwari et al (1997). 4 The TIGR website http://www.tigr.org/softlab/glimmer/glimmer.…”

“…The basic algorithm used in GeneScan 2 has already been described in detail (Tiwari et al 1997). The essential feature of this program is its detection of the 3-base periodicity which is shown by coding regions to identify putative genes.…”

“…In the former category, there are two methods that are based on the correlation properties of DNA sequences, GeneScan (Tiwari et al 1997) and the Coding Region Finder (Ossadnik et al 1994) methods. The latter category includes SELFID (Audic and Claverie 1998) and GLIMMER (Delcher et al 1999), which are interpolated Markov model based techniques.…”

Prediction of Protein Coding Regions in DNA Sequences Using Fourier Spectral Characteristics

“…Existing Discrete Fourier transform (DFT)-based algorithms for identifying protein coding regions in DNA sequences [9,2,3,7] …”