1 / 55

Lecture II: Genomic Methods

Lecture II: Genomic Methods. Dennis P. Wall, PhD Frederick G. Barr, MD, PhD Deborah G.B. Leonard, MD, PhD. Why Pathologists? We have access, we know testing. Personalized Risk Prediction, Medication Dosing, Diagnosis/ Prognosis. Pathologists. Access to patient’s genome.

elias
Download Presentation

Lecture II: Genomic Methods

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Lecture II: Genomic Methods Dennis P. Wall, PhD Frederick G. Barr, MD, PhD Deborah G.B. Leonard, MD, PhD TRiG Curriculum: Lecture 2

  2. Why Pathologists? We have access, we know testing Personalized Risk Prediction, Medication Dosing, Diagnosis/ Prognosis Pathologists Access to patient’s genome Physician sends sample to Pathology (blood/tissue) Just another laboratory test TRiG Curriculum: Lecture 2

  3. The path to genomic medicine Sample Collection Pathologists Sample Collection Testing: Sequencing, Gene chips Access to patient’s genome Analysis TRiG Curriculum: Lecture 2

  4. What we will cover today: • Types of genetic alterations • Current and future molecular testing methods • Cytogenetics, in situ hybridization, PCR • Microarrays • Genotyping • Expression profiling • Copy number variation • Next generation sequencing (NGS) • Whole genome • Transcriptome TRiG Curriculum: Lecture 2

  5. DNA alterations – the small stuff Point mutation CCTGAGGAG CCTGTGGAG Example: hemoglobin, beta – sickle cell disease Deletion/Insertion GAATTAAGAGAAGCA GAAGCA Example: epidermal growth factor receptor – lung cancer Repeat alteration TTCCAG…(CAG)5…CAGCAA TTCCAG…(CAG)60…CAGCAA Example: huntingtin – Huntington disease TRiG Curriculum: Lecture 2

  6. DNA alterations – the bigger stuff Deletion/ Insertion Example: 22q11.2 region – DiGeorge syndrome Example: 17q21.1 (ERBB2) – Breast cancer Amplification Example: t(11;22)(q24;q12) – Ewing’s sarcoma Translocation Der 22 22 11 Der 11 TRiG Curriculum: Lecture 2

  7. Current strategies to detect DNA alterations Cytogenetics: Large indels, amplification, translocations In situ hybridization: large indels, amplification, translocations t(6;15) in a woman with repeated abortions EGFR amplification in glioblastoma http://www.indianmedguru.com http://moon.ouhsc.edu TRiG Curriculum: Lecture 2

  8. Current strategies to detect DNA alterations PCR-based approaches: Mutations, small indels, repeat alterations, large indels, amplification, translocations Factor V Leiden mutation Alsmadi OA, et al. BMC Genomics 2003 4:21 TRiG Curriculum: Lecture 2

  9. What we will cover today: • Types of genetic alterations • Current and future molecular testing methods • Cytogenetics, in situ hybridization, PCR • Microarrays • Genotyping • Expression profiling • Copy number variation • Next generation sequencing (NGS) • Whole genome • Transcriptome TRiG Curriculum: Lecture 2

  10. DNA microarray - the basics • Purpose: multiple simultaneous measurements by hybridization of labeled probe • DNA elements may be: • Oligonucleotides • cDNA’s • Large insert genomic clones • Microarray is generated by: • Printing • Synthesis TRiG Curriculum: Lecture 2

  11. Microarray technologies TRiG Curriculum: Lecture 2

  12. Organization of a DNA microarray 1.28 cm 1.28 cm (adapted from Affymetrix) TRiG Curriculum: Lecture 2

  13. Hybridization of a labeled probeto the microarray (adapted from Affymetrix) TRiG Curriculum: Lecture 2

  14. Detection of hybridization on microarray Light from laser (adapted from Affymetrix) TRiG Curriculum: Lecture 2

  15. Hybridization intensities on DNA microarray following laser scanning TRiG Curriculum: Lecture 2

  16. Overview of SNP array technology LaFramboise T. Nucleic Acids Res. 2009; 37:4181 TRiG Curriculum: Lecture 2

  17. Microarray Applications • DNA analysis • Polymorphism/mutation detection – e.g. Disease susceptibility testing Drug efficacy/sensitivity testing • Copy number detection (comparative genomic hybridization) – e.g. Constitutional or cancer karyotyping • Bacterial DNA – e.g. Identification and speciation • RNA analysis • Expression profiling – e.g. Breast cancer prognosis Cancer of unknown primary origin cv TRiG Curriculum: Lecture 2

  18. Genome-wide association studies of lung cancer microarray with 317,139 SNP’s Cases/controls From different populations Hung RJ, et al. Nature Genetics. 2008; 452:633 TRiG Curriculum: Lecture 2

  19. Genotype calling Hybridization intensities translated into genotypes Large SNP numbers requires automated procedure Recent algorithms – clustering/pooling strategies • Raw hybridization intensities normalized • Information combined across different samples at each SNP • Assign genotypes to entire clusters • For each sample, estimate probability of each of three genotype calls at each SNP • Genotype assigned based on defined threshold of probability • Missing genotypes dependent on algorithm & threshold used Teo YY, Curr Op in Lipidology. 2008; 19:133 TRiG Curriculum: Lecture 2

  20. Genotyping - Limitations & quality control • Accuracy of algorithm • Depends on number of samples in each cluster • Prone to errors for small number of samples or SNP’s with rare alleles • High rates of missing genotypes: • Array problems – plating/synthesis issue • Poor quality DNA – degradation • Hybridization failure • Differential performance between SNP’s • Excess heterozygosity - sample contamination? Just another laboratory test TRiG Curriculum: Lecture 2

  21. Analyzed 8,101 genes on chip microarrays • Reference= pooled cell lines • Breast cancer subgroups Perou CM, et al. Nature. 2000; 406, 747 TRiG Curriculum: Lecture 2

  22. Original two probe strategy for expression profiling on cDNA arrays Duggan DJ, et al., Nature Genetics. 1999; 21:10 TRiG Curriculum: Lecture 2

  23. Expression profiling: challenges and limitations Biological • Dynamic & complex nature of gene expression • Heterogeneous nature of tissue samples • Variation in RNA quality Technological • Reproducibility across microarray platforms • Selection of probes – dependence on binding efficiency • Controlling for technical variability Statistical/bioinformatic • Adequate experimental design • Normalization to remove variability among chips • Multiple testing correction • Validation of results Just another laboratory test TRiG Curriculum: Lecture 2

  24. Copy number variation: Comparative genomic hybridization Tumor DNA Reference DNA Hybridization CGH Array-CGH Arrayed DNA’s Metaphase Chromosomes Deletion Gain Deletion Gain http://www.advalytix.com/advalytix/hybridization_330.htm TRiG Curriculum: Lecture 2

  25. Constitutional genomic imbalances detected by copy number arrays 10.9 Mb deletion at 7q11 7.2 Mb duplication on 11q Miller DT, et al, Amer J Hum Genet. 2010; 86:749 TRiG Curriculum: Lecture 2

  26. Copy number - Limitations & quality control Artifacts may be caused by: • GC content • Wavy patterns correlate with GC content • Algorithms developed to remove waviness • DNA sample quantity and quality • Can impact on level of signal noise and false positive rate • Whole genome amplification associated with signal noise • Sample composition • In cancer studies, normal cells dilute cancer aberrations • Tumor heterogeneity will also affect copy number Just another laboratory test TRiG Curriculum: Lecture 2

  27. What we will cover today: • Types of genetic alterations • Current and future genetic test methods • Cytogenetics, in situ hybridization, PCR • Microarrays • Genotyping • Expression profiling • Copy number variation • Next generation sequencing (NGS) • Whole genome • Transcriptome TRiG Curriculum: Lecture 2

  28. Cancer Treatment: NGS in AML Welch JS, et al. JAMA, 2011;305, 1577 TRiG Curriculum: Lecture 2

  29. Case History 39 year old female with APML by morphology Cytogenetics and RT-PCR unable to detect PML-RAR fusion Clinical question: Treat with ATRA versus allogeneic stem cell transplant TRiG Curriculum: Lecture 2

  30. Methods/Results Paired-end NGS sequencing Result: Cytogenetically cryptic event: novel fusion protein Took 7 weeks TRiG Curriculum: Lecture 2

  31. 77-kilobase segment from Chr. 15 was inserted en bloc into the second intron of the gene RARA on Chr. 17. TRiG Curriculum: Lecture 2

  32. Workflow Raw Data Analysis • Image processing and base calling Whole Genome Mapping • Alignment to reference genome Variant Calling • Detection of genetic variation • (SNPs, Indels, Insertions) Annotation Linking variants to biological information TRiG Curriculum: Lecture 2

  33. Overview of Paired End Sequencing Short Insert Annealed to Surface Synthesized Adapters Ligated Sequenced Random Shearing DNA Sequencing done with labeled NTPs and massively parallel TRiG Curriculum: Lecture 2

  34. Short read output format Read ID Sequence Quality line TRiG Curriculum: Lecture 2

  35. Quality control is critical Just another laboratory test TRiG Curriculum: Lecture 2

  36. Measuring Accuracy • Phred is a program that assigns a quality score to each base in a sequence. These scores can then be used to trim bad data from the ends, and to determine how good an overlap actually is. • Phredscores are logarithmically related to the probability of an error: a score of 10 means a 10% error probability; 20 means a 1% chance, 30 means a 0.1% chance, etc. • A score of 20 is generally considered the minimum acceptable score. TRiG Curriculum: Lecture 2

  37. Workflow Raw Data Analysis • Image processing and base calling Whole Genome Mapping • Alignment to reference genome Variant Calling • Detection of genetic variation • (SNPs, Indels, Insertions) Annotation Linking variants to biological information TRiG Curriculum: Lecture 2

  38. Alignment/Mapping GGTATAC… …CCATAG TATGCGCCC CGGAAATTT CGGTATAC CGGTATAC …CCAT CTATATGCG TCGGAAATT GCGGTATA GGCTATATG CTATCGGAAA …CCAT TTGCGGTA C… …CCA AGGCTATAT CCTATCGGA C… TTTGCGGT AGGCTATAT …CCA GCCCTATCG ATAC… AAATTTGC AGGCTATAT …CC GCCCTATCG …CC TAGGCTATA GCGCCCTA AAATTTGC GTATAC… …CCATAGGCTATATGCGCCCTATCGGCAATTTGCGGTATAC… GAAATTTGC GGAAATTTG CGGAAATTT CGGAAATTT TCGGAAATT CTATCGGAAA CCTATCGGA TTTGCGGT GCCCTATCG AAATTTGC GCCCTATCG AAATTTGC …CC ATAC… …CCATAGGCTATATGCGCCCTATCGGCAATTTGCGGTATAC… Read depth is critical for accurate reconstruction TRiG Curriculum: Lecture 2

  39. Alignment approaches KoboldtDC, et al. Brief Bioinform 2010 Sep;11(5):484-98 TRiG Curriculum: Lecture 2

  40. Short read alignment • Given a reference and a set of reads, report at least one “good” local alignment for each read if one exists • Approximate answer to question: where in genome did read originate? • What is “good”? For now, we concentrate on: • Fewer mismatches = better • Failing to align a low-quality base is better than failing to align a high-quality base …TGATCATA… GATCAA …TGATCATA… GAGAAT better than …TGATATTA… GATcaT …TGATCATA… GTACAT better than TRiG Curriculum: Lecture 2

  41. Post alignment: what do you get? Alignment of reads including read pairs CIGAR field SAM file Read Pair Simplified pileup output Li H, et al. Bioinformatics. 2009;25:2078 TRiG Curriculum: Lecture 2

  42. Workflow Raw Data Analysis • Image processing and base calling Whole Genome Mapping • Alignment to reference genome Variant Calling • Detection of genetic variation • (SNPs, Indels, Insertions) Annotation Linking variants to biological information TRiG Curriculum: Lecture 2

  43. Discovering Genetic Variation ATCCTGATTCGGTGAACGTTATCGACGATCCGATCGA ATCCTGATTCGGTGAACGTTATCGACGATCCGATCGA CGGTGAACGTTATCGACGATCCGATCGAACTGTCAGC GGTGAACGTTATCGACGTTCCGATCGAACTGTCAGCG TGAACGTTATCGACGTTCCGATCGAACTGTCAGCGGC TGAACGTTATCGACGTTCCGATCGAACTGTCAGCGGC TGAACGTTATCGACGTTCCGATCGAACTGTCAGCGGC GTTATCGACGATCCGATCGAACTGTCAGCGGCAAGCT TTATCGACGATCCGATCGAACTGTCAGCGGCAAGCT ATCCTGATTCGGTGAACGTTATCGACGATCCGATCGAACTGTCAGCGGCAAGCTGATCGATCGATCGATGCTAGTG reference genome TTATCGACGATCCGATCGAACTGTCAGCGGCAAGCT TCGACGATCCGATCGAACTGTCAGCGGCAAGCTGAT ATCCGATCGAACTGTCAGCGGCAAGCTGATCG CGAT TCCGATCGAACTGTCAGCGGCAAGCTGATCG CGATC TCCGATCGAACTGTCAGCGGCAAGCTGATCGATCGA GATCGAACTGTCAGCGGCAAGCTGATCG CGATCGA AACTGTCAGCGGCAAGCTGATCG CGATCGATGCTA TGTCAGCGGCAAGCTGATCGATCGATCGATGCTAG TCAGCGGCAAGCTGATCGATCGATCGATGCTAGTG INDELs SNPs TRiG Curriculum: Lecture 2

  44. TRiG Curriculum: Lecture 2

  45. TRiG Curriculum: Lecture 2

  46. Workflow Raw Data Analysis • Image processing and base calling Whole Genome Mapping • Alignment to reference genome Variant Calling • Detection of genetic variation • (SNPs, Indels, Insertions) Annotation Linking variants to biological information TRiG Curriculum: Lecture 2

  47. Where to go to annotate genomic data, determine clinical relevance? Online Mendelian Inheritance in Man(http://www.ncbi.nlm.nih.gov/omim) International HapMap project (http://hapmap.ncbi.nlm.nih.gov) Human genome mutation database (http://www.hgvs.org/dblist/glsdb.html) PharmGKB(http://www.pharmgkb.org) Scientific literature TRiG Curriculum: Lecture 2

  48. Case-control study design = variable results • Need for Clinical Grade Database • Ease of use • Continually updated • Clinically relevant SNPs/variations Ng PC, et al. Nature. 2009; 461: 724 TRiG Curriculum: Lecture 2

  49. Cancer Treatment: NGS of Tumor Jones SJM, et al. Genome Biol. 2010;11:R82. TRiG Curriculum: Lecture 2

  50. Case History • 78 year old male • Poorly differentiated papillary adenocarcinoma of tongue • Metastatic to lymph nodes • Failed chemotherapy • Decision to use next-generation sequencing methods TRiG Curriculum: Lecture 2

More Related