基因组测序的原理与方法课件.ppt

走向人类赖以生存的物质基础 抗病、抗虫和抗极端环境GM农作物 高生殖率、高生长率、高营养率的GM家畜、家禽和水产品新品种 维生素和营养物质富集的水果和蔬菜 生物杀虫剂、除草剂和抗病药物 微生态环境下生产的有机食品 * Data assessment I – Read quality distribution Low Quality ? High Quality Trim: 3’ end trim if QN 20 Filter: Percent (hight quality Q 30) 60 Assessment: Distance Distrubition between two Low quality (Q20) 454 dinucleotide proportion check 454 raw reads quality Data assessment II – Library insert size Numbers of reads with non-insert DNA (full length adapter) in different insert size libraries Data assessment III – Mapping Rate Solexa Sequencing Data Usage in 500bp Library Data assessment IV – Duplication assessment Duplicates detection and filter F R N N 2N Qaverage 20 ? Lane data usage in different solexa library- Fiter duplication reads Average Reads per StartPoint Read CorrectionCorrect Illumina GA short reads Kmer = 17 Genome Size Prediction: M = N * ( L-K+1)/L N = Total Length (bp) /Genome size L= Average Rads Length (bp) M Genome size estimation using Kmer Before estimating the genome size, we set a hypothesis: the k-mer we picked out from the genome can ergodic the whole genome sequence.According to the Lander waterman algorithm, the algorithm should be represented as: G= Knum / Kdepth Here, G is the genome size, Knum is the total number of k-mer and Kdepth is the expected depth of the k-mer. If we obtain the expected depth of k-mer, we can calculate the genome size. Because the distribution of k-mer frequency yields to Poisson distribution, we can consider the peak of the k-mer distribution curve as the expected depth of k-mer and calculate the genome size. Note: A total of 15,437,084,746 Kmers, the peak value on the right figure is 8, so the genome size is estimated as: 15,437,084,746/8=1.93G High Quality Read Rate after preprocess Assembly: Raw data VS preprocessed Data ? Questions Genome size estimation methods (K-mer Cov) Assembly optimization (parameters) Assembly evaluation (454_Solexa