Bioinformatics and Data Analysis

Transcription

1 University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln Core for Applied Genomics and Ecology (CAGE) Food Science and Technology Department Bioinformatics and Data Analysis Follow this and additional works at: Part of the Food Microbiology Commons "Bioinformatics and Data Analysis" (2008). Core for Applied Genomics and Ecology (CAGE) This Article is brought to you for free and open access by the Food Science and Technology Department at DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in Core for Applied Genomics and Ecology (CAGE) by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln.

2 Presentation from the workshop: Next-gen Sequencing for Research Scientists September 15-16, 2008 East Campus Union Presented by: Department of Food Science & Technology, University of Nebraska Lincoln

3 The UNL Next Generation Sequencing Workshop 2008 Powered by

4 Bioinformatics and data analysis

5 Run-time applications Post-run applications Image processing Signal processing de novo Assembly Reference Mapper Variant analysis

6 Run-time applications

7 Post-run applications

8 Post-run applications

9 Post-run applications

10 Off-rig Dual processer File server Internet On-Rig computer 4-node Linux machine

11 Image processing Signal processing Current data processing pipeline Off-rig Dual processer Linux File server File storage Internet-workstations On-Rig computer 4-node Linux machine Post-run applications

12 Storage of image data files is only temporary: Each run generates about 12Gb of image data Only 10 runs stored on the On-rig computer Can only store another 10-runs of complete Data sets on file-server

13

14

15 Image Processing On rig

16

17 Quality filtering and trimming 1. After signal normalization and removal of problematic wells 2. Each read run through five different quality filters 3. Filtering occurs in flow space as opposed to nucleotide space 4. Summary of filtering results output as series of metrics files 5. Two levels of filters Read-rejection filters 1. Keypass filter 2. Dot filter 3. Mixed filter --Must pass all three filters otherwise rejected Read-trimming filters 1. Signal intensity filter (for 3 end of read) 2. Primer filter --last two filters trim reads back from 3 end and if trim leaves less than 85 bases, read is rejected

18 Read-rejection filters Keypass filter verifies that sequences in a well has the TCAG key sequence Dot filter dots are instances of 4 successive flows with NO signal Generation. If the last flow before the dot occurs before 85 bases, Entire read is rejected. If dot occurs after that, trims to first Dot occurrence Mixed filter two phase filter --phase one ensures that less than 70% of flows have signal --phase two looks at intensity of positive flows, if more than 30% of reads are in middle ground (e.g. 0.5, 1.5, 2.5, etc.) in intensity, then read is rejected

19 Read trimming filters Signal intensity filter two-phase filter --phase one defines range of flow signals from 0-1 and considers intensity from borderline. If more than 30% of flows in a read fall in the borderline region, trim back and if trim generates sequences <85, reject --phase two is a Valley filter uses all reads passing other filters to measure valleys between 0-1mer, 1mer-2mer, 2mer-3mer peaks then scans reads counting number of flow signals occurring within 0.1 value of the valley minima Calculating the ratio: valley flows over entire read valley flows in test High valley filter values trims reads back to last base before Valley that exceeds threshold in read

20 Read-trimming filters Primer filter scans ends of reads for Adapter sequences (A and B) and Trims from reads If any trimmed reads are <84 flows rejected

21 Individual base quality scores Uses metrics similar to Phred compares Flowgram signal characteristics to those from Known high-quality reads Uses training sets of the reads from the run Performs multivariate analysis to determine set of properties best fitting bins of basecalls Associates training set accuracy rates Of basecalls in each bin to the Scale Q = -10 log 10 (accuracy) Trace properties of each base are then determine The bin into which it falls

22 Flow characteristics used to measure quality

23 Output files from signal processing

24 Data analysis: Assembly and Mapping

25

26

27

28 Default parameters

29

30

31

32

33

34 Start button Becomes live When files are added

35

36

37

38

39 Highlighting a read And right clicking Allows you to go back to Flowgram from that read

40 Zooming into Section of a flowgram

41 Reference Mapper

42

43

44

45

46

47

48

49

50

51 The amplicon variant analyzer (AVA)

52 Allows you to add reads and associate them with A specific segment or locus

53

54

55

56

57

58

59

60 Microbial Community Analysis

61

62 Save file to Local directory

63

64

65

66