(README.TXT)

Software: DNAprobe

Supplied files:
readme.txt (this file)
dnaprobe.exe (the software!)

Purpose:
To isolate genes for new members of a protein family, it is frequently
necessary to design oligonucleotides on the basis of protein alignments. 
DNAPROBE assists in the design of oligonucleotides in such cases.


Installation:
Copy DNAprobe.exe to a directory of your choice and run it!


Status:
Freeware.


Distribution:
Freely distributable, but all distributions must include this readme.txt file.
Not to be sold or hired.
If you find DNAprobe useful, please feel free to acknowledge the authors, John
and Martin.


Contact Information:
Please address queries to martin.drummond@bbsrc.ac.uk
(Software author: John Stamper)


Uploaded to Simtel.net by JIC Computing Services on behalf of John Stamper and
Martin Drummond.




     To isolate genes for new members of a protein family, it is frequently
necessary to design oligonucleotides on the basis of protein alignments. 
DNAPROBE assists in the design of oligonucleotides in such cases. It uses MSF
protein alignments generated by the program PILEUP and codon frequency tables
from the Wisconsin (GCG) sequence analysis software package. The basic
algorithm calculates the most probable nucleotide sequence corresponding to the
alignment by summing the products of the frequency with which a particular amino
acid occurs at any one position and the codon usage frequency for that amino acid.
Where an alignment contains a cluster of closely related sequences, the user may
choose to implement a weighting option that reduces their individual contributions
to the output in order that they do not overwhelm the contributions of more distantly
related sequences which are under-represented. This may be done in two ways; (a)
using the algorithm of Gerstein et al. (J. Mol. Biol., 236, 1067-1078) and the
information contained in the dendrogram file written by PILEUP,  or (b) by
specifying weights in a separate file which is then read by the program, an option
which enables one to give increased weight to sequences known to be more closely
related to the target organism.
     The user specifies the length of the oligonucleotide required, and the program
then scans the most probable nucleotide sequence to find segments of this length
having the greatest overall probability of matching the sequence of an unknown gene
of the desired class. Users will generally wish to exclude probes derived from parts
of the alignment that contains gaps, and the program contains an option to do so. A
chosen number of the best oligonucleotide segments are listed, together with their 
matching probability, their melting temperatures according to the algorithm of Freier
et al. ( Proc. Natl. Acad. Sci. USA 83, 9373-9377.), and their positions in the
sequence. The probability of occurrence of the preferred base at each position in the
sequence can be displayed graphically, as well as the cumulative probability for
consecutive oligonucleotides over the entire sequence. The program displays the
most probable nucleotide sequence together with its translation, which is also
aligned to the input protein sequences. This hypothetical translation product is useful
for matching oligonucleotides to the input protein sequences, although occasionally
it contains amino acids not present in the original alignment as a result of different
forces predominating at different positions within a codon; this is not a cause for
concern.
     The efficiency of both PCR and filter hybridisation can be enhanced by using
inosine or mixtures of bases at positions for which no strong prediction can be
made, and to facilitate this kind of choice, DNAPROBE contains an option to
display the probability of occurrence for each base at every position of a chosen
oligonucleotide. 
     The program is written in C ++ and runs under Windows.

