Документ взят из кэша поисковой машины. Адрес оригинального документа : http://mccmb.belozersky.msu.ru/2013/abstracts/abstracts/117.pdf Дата изменения: Mon Mar 18 14:50:02 2013 Дата индексирования: Thu Feb 27 20:54:22 2014 Кодировка:

Abstract to `Geno me Disorder' Detection of some new statistical biases in next generation sequencing, Illumina platform.

Irina Abnizova, Steve Leonard, Marina Gourtovaia, Fedor Naumenko, Rene te Boekhorst and David Jackson Wellcome Trust Sanger Institute, Hinxton, Cambridge ,UK

A rare mutation detection is one of the most powerful applicat ions o f next generat ion sequencing (NGS). Therefore the fidelit y o f NGS laboratory, machine and post-processing informat ics becomes increasingly crit ical. Here we describe a new statist ical biases in high qualit y mis match detection o f NGS, Illumina platform. Some o f these biases are already known to have dramatic effect on downstream SNP analysis. So me arrived recent ly, and related to PCR and library preparat ion errors. We introduce novel informat ics methods to confident ly filter these high qualit y artefacts fro m sequencing data. We also demonstrate several new sequence qualit y measures o f biases' detection. These artefacts are difficult (if possible) to see with convent ional qualit y measures.