唐主任-Miseq应用初步经验.pptVIP

If we removed those duplication and reads with low MQ, the result should be ref/ref only 4 reads with G covered in the site, it is more likely to be a hom instead of het. Miseq automatic pipeline did not remove duplication and used bad MQ reads for genotyping. By contrast, if we remove duplicates and used reads with MQ50 all false positive was no longer exist results matched with those from Sanger sequencing If we removed duplication but not MQ filtering, the number for G dropped from 54 to 23, indicating 31 PCR duplications were recognized,. If we only do MQ filtering (keep MQ50), only 8 for G left, but 4 of them are duplications in this example. C, T was moved due to low MQ. Not sufficient coverage! Outlines Data Producing Selection of targeted panel Preparation of libraries Sequencing inside the MiSeq Data Processing (HKU) Computing analysis Data Analysis bioinformatics Future endeavor Benign by ClinVar Pathogenic by ClinVar c.6380-12_c.6380-13insT Data Producing Selection of targeted panel Preparation of libraries Sequencing inside the MiSeq Data Processing (WL Lab) Computing analysis Data Analysis bioinformatics Based on fastq file, total produced 1.2 Gb data from 12 samples ! Data Producing Selection of targeted panel Preparation of libraries Sequencing inside the MiSeq Data Processing (CGS ?) Computing analysis Data Analysis bioinformatics Future Endeavor Future Endeavor Short term Targeted panel Marfan, NF, DMD RP Hypothyroid Mid term Exome sequencing ARSCA Long term Accreditation Bioinformatic team MiSeq Implementation in CGS experience sharing Tong Ming For 16 August 2014 Outlines Data Producing Selection of targeted panel Preparation of libraries Sequencing inside the MiSeq Data Processing (WL Lab) Computing analysis Data Analysis bioinformatics Future endeavor Outlines Data Producing Selection of targeted panel Preparation of libraries Sequencing inside the MiSeq Data Processing (WL / HKU) Computing analysis Data Analysis bioinform