A DOT-MATRIX PROGRAM WITH DYNAMIC THRESHOLD CONTROL SUITED FOR GENOMIC DNA AND PROTEIN-SEQUENCE ANALYSIS

Citation
Ell. Sonnhammer et R. Durbin, A DOT-MATRIX PROGRAM WITH DYNAMIC THRESHOLD CONTROL SUITED FOR GENOMIC DNA AND PROTEIN-SEQUENCE ANALYSIS, Gene, 167, 1995, pp. 1-10
Citations number
35
Categorie Soggetti
Biology
Journal title
GeneACNP
ISSN journal
03781119
Volume
167
Year of publication
1995
Pages
1 - 10
Database
ISI
SICI code
0378-1119(1995)167:<1:ADPWDT>2.0.ZU;2-O
Abstract
Graphical dot-matrix plots can provide the most complete and detailed comparison of two sequences. Presented here is DOTTER(2), a dot-plot p rogram for X-windows which can compare DNA or protein sequences, and a lso DNA versus protein. The main novel feature of DOTTER is that the u ser can vary the stringency cutoffs interactively, so that the dot-mat rix only needs to be calculated once. This is possible thanks to a 'Gr eyramp tool' that was developed to change the displayed stringency of the matrix by dynamically changing the greyscale rendering of the dots . The Greyramp tool allows the user to interactively change the lower and upper score limit for the greyscale rendering. This allows explora tion of the separation between signal and noise, and fine-grained visu alisation of different score levels in the dot-matrix. Other useful fe atures are dot-matrix compression, mouse-controlled zooming, sequence alignment display and saving/loading of dot-matrices. Since the matrix only has to be calculated once and since the algorithm is fast and li near in space, DOTTER is practical to use even for sequences as long a s cosmids. DOTTER was integrated in the gene-modelling module of the g enomic database system ACEDB(3). This was done via the homology viewer BLIXEM in a way that also allows segments from the BLAST suite of sea rching programs to be superimposed on top of the full dot-matrix. This feature can also be used for very quick finding of the strongest matc hes. As examples, we analyse a Caenorhabditis elegans cosmid with seve ral tandem repeat families, and illustrate how DOTTER can improve gene modelling.