%0 Journal Article %1 Zhang:2005aa %A Zhang, Yu %A Waterman, Michael S %D 2005 %J Proc Natl Acad Sci USA %K Acid Acid, Alignment, Base DNA, Data, Genetic, Homology, Models, Molecular Nucleic Repetitive Sequence Sequence, Sequences, %N 5 %P 1285--90 %R 10.1073/pnas.0409240102 %T An Eulerian path approach to local multiple alignment for DNA sequences %V 102 %X Expensive computation in handling a large number of sequences limits the application of local multiple sequence alignment. We present an Eulerian path approach to local multiple alignment for DNA sequences. The computational time and memory usage of this approach is approximately linear to the total size of sequences analyzed; hence, it can handle thousands of sequences or millions of letters simultaneously. By constructing a De Bruijn graph, most of the conserved segments are amplified as heavy Eulerian paths in the graph, and the original patterns distributed in sequences are recovered even if they do not exist in any single sequence. This approach can accurately detect unknown conserved regions, for both short and long, conserved and degenerate patterns. We further present a Poisson heuristic to estimate the significance of a local multiple alignment. The performance of our method is demonstrated by finding Alu repeats in the human genome. We compare the results with Alus marked by repeatmasker, where the two programs are in good agreement. Our method is robust under various conditions and superior to other methods in terms of efficiency and accuracy.

@article{Zhang:2005aa, abstract = {Expensive computation in handling a large number of sequences limits the application of local multiple sequence alignment. We present an Eulerian path approach to local multiple alignment for DNA sequences. The computational time and memory usage of this approach is approximately linear to the total size of sequences analyzed; hence, it can handle thousands of sequences or millions of letters simultaneously. By constructing a De Bruijn graph, most of the conserved segments are amplified as heavy Eulerian paths in the graph, and the original patterns distributed in sequences are recovered even if they do not exist in any single sequence. This approach can accurately detect unknown conserved regions, for both short and long, conserved and degenerate patterns. We further present a Poisson heuristic to estimate the significance of a local multiple alignment. The performance of our method is demonstrated by finding Alu repeats in the human genome. We compare the results with Alus marked by repeatmasker, where the two programs are in good agreement. Our method is robust under various conditions and superior to other methods in terms of efficiency and accuracy.}, added-at = {2007-09-17T20:19:41.000+0200}, affiliation = {Department of Mathematics, University of Southern California, 1042 West 36th Place, DRB289, Los Angeles, CA 90089-1113, USA.}, author = {Zhang, Yu and Waterman, Michael S}, biburl = {https://www.bibsonomy.org/bibtex/2094f2ea8388f71a5fb448d2e3517059b/dzerbino}, doi = {10.1073/pnas.0409240102}, interhash = {34c603ae0c0a986e8eada54cfaa160ee}, intrahash = {094f2ea8388f71a5fb448d2e3517059b}, issue = {5}, journal = {Proc Natl Acad Sci USA}, keywords = {Acid Acid, Alignment, Base DNA, Data, Genetic, Homology, Models, Molecular Nucleic Repetitive Sequence Sequence, Sequences,}, language = {English}, local-url = {file://localhost/Users/danielzerbino/Documents/Papers/2005/Zhang/Proc%20Natl%20Acad%20Sci%20USA%202005%20Zhang.pdf}, month = Feb, number = 5, pages = {1285--90}, pii = {0409240102}, pmid = {15668398}, timestamp = {2007-09-17T20:19:43.000+0200}, title = {An Eulerian path approach to local multiple alignment for DNA sequences}, uri = {papers://055852FE-1648-42FE-91D0-8CA474D2B905/Paper/p7}, volume = 102, year = 2005 }