Computational methods for transcription factor binding site analysis
OwnerMichael F. Chou, David M. Geshwindinventors
Life of this patent
Estimating the term…
Abstract
The invention is a new method to computationally analyze nucleic acid-protein binding data in a systematic or statistical manner in order to determine the DNA binding sequence for proteins which bind to DNA and are usually involved in the regulation of the expression of genes, and may enhance, promote or repress the expression of the gene.
Claims26
A method for determining a set, A, of double-stranded DNA sequences equal to or less than length L that are potential binding sites of a DNA-binding protein which has been exposed to a set of nucleic acid probes P, wherein jj is an index of said set P, subset B is the subset of probes of said set P to which said protein has bound, set T is the set of all possible nucleotide sequences of non-zero length and shorter or equal to length L and ii is an index of said set T, and where said set A is in an initially empty state, for each possible nucleotide sequence Tii, wherein set Hii is a corresponding initially empty set, the following steps are performed utilizing, at least in part, a digital computing device:
- a. comparing said sequence Tii to the sequence of each member, Pjj, of set P and:
- b1. if Tii is a subsequence of the sequence of Pjj;
- then,
- b2. adding Pjj to a set Hii of hypothetical probes corresponding to sequence Tii;
- c. determining if Hii is a subset of subset B;
- and,
- d. if the condition of step c is true, adding sequence Tii to said set A.
A method utilizing, at least in part, a digital computing device for determining a set of potential double-stranded DNA binding site sequences equal to or less than a probe length L, of a DNA-binding protein which has been exposed to a set of nucleic acid probes P comprising sequences of length L, and wherein subset B is the subset of probes of said set P to which said protein has bound, by performing multiple sequence alignment analysis on the sequences of subset B.
Dependencies are inferred from the claim text.