findIR is a MATLAB-based tool for accurate detection of perfect inverted repeats in DNA sequences. It is published in Bioinformatics 30:887-888.
You can examine the supplemental tables and figures for our findIR publication:SupplementalTableFigure.pdf
- You can download our source code for this program: SourceCodeData
- For how to use our program, please read the file: readme.txt
- For how to obtain the large test data for testing, please read the file: testcase_info.txt
For general users, we have created a new, better tool called detectIR. detectIR can detect not only perfect inverted repeats as findIR does, but also imperfect inverted repeats that findIR cannot detect. In terms of detecting perfect inverted repeats, detectIR has the same accuracy as findIR because both tools adopt exhaustive search strategy, but it is much faster because detectIR takes advantage of MATALB vector calculation power whereas findIR does not.
However, we do believe that findIR still has its scientific value, because it is the first work that utilizes a prime number-based algorithm to replace conventional string search and comparison algorithms in nucleotide sequence pattern search. We hope that this idea can trigger more innovative ideas in computational biology that utilize calculation, search and comparison of numbers (e.g., primer number, complex number and so on) into DNA and RNA pattern search, which might be more accurate and efficient than conventional algorithms of string search and comparison. Indeed, findIR has inspired us to have another novel idea that uses a complex number-based algorithm for nucleotide sequence pattern search, as implemented in detectIR.