PILER: identification and classification of genomic repeats
PILER: identification and classification of genomic repeats
复制标题
DOI：
10.1093/bioinformatics/bti1003
复制
发表时间：
2005-06-01
期刊：
影响因子：
5.8
通讯作者：
Myers, EW
中科院分区：
文献类型：
--
作者：
Edgar, RC;Myers, EW
在真核生物基因组中，重复元件如随体和转座子是普遍存在的。从头计算识别和分类的这些元素是一个具有挑战性的问题。因此，测序基因组的重复注释在历史上主要依赖于与已知重复家族的手工库的序列相似性。我们提出了一种新的方法从头重复注释，利用特定类别的重复引起的局部对齐的特征模式。我们描述PILER，一个包的有效的搜索算法，用于识别这种模式。使用PILER发现的新的重复报道了智人，拟南芥和果蝇。
Repeated elements such as satellites and transposons are ubiquitous in eukaryotic genomes. De novo computational identification and classification of such elements is a challenging problem. Therefore, repeat annotation of sequenced genomes has historically largely relied on sequence similarity to hand-curated libraries of known repeat families. We present a new approach to de novo repeat annotation that exploits characteristic patterns of local alignments induced by certain classes of repeats. We describe PILER, a package of efficient search algorithms for identifying such patterns. Novel repeats found using PILER are reported for Homo sapiens, Arabidopsis thalania and Drosophila melanogaster.