Preparation of Sample Hybridization Scanning and Image Analysis

Transcription

1 Preparation of Sample Hybridization Scanning and Image Analysis

2 Sample preparation 1. Design experiment 2. Perform experiment Question? Replicates? Test? mutant wild type 3. Precipitate RNA 4. Label RNA Eukaryote/prokaryote? Cell wall? Amplification? Direct or indirect? Label?

3 Sample Preparation - Check quality of RNA Load ng of RNA on the BioAnalyzer eukaryotic: 28S/18S 2.0 prokaryotic: 23S/16S 2.0

4 Sample preparation AAAAAAAAA TTTTTTTTT RNA cdna aa-dutp Revese Transcriptase (RT) oligo dt primer nucleotides (dntp) amino-allyl-dutp (aa-dutp)

5 Sample preparation AAAAAAAAA TTTTTTTTT RNA cdna Hydrolysis (NaOH and EDTA Æ Tris) Clean-up

6 Sample preparation TTTTTTTTT cdna Coupling (Cy3 or Cy5) Quenching Clean-up

7 Sample preparation TTTTTTTTT TTTTTTTTT Combine Hybridize

8 Hybridization

9 Sample Preparation - Eberwine SAMPLE 42 C 2 h + Reverse Transcriptase RNA ssdna T7 70 C 10 min 16 C 2 h + RNase H + Polymerase T7 pol clean up dsdna 37 C 6 h dsdna arna + Biotin-labeled nucleotides

10 Detection of Biotin (Affymetrix) Streptavidin Phycoerythrim = SAPE ( ) anti-sape IgG biotinylated anti-anti IgG

11 Materials and Methods All protocols can be found here: and here:

12 Scanning and image analysis by H. Bjørn Nielsen Scanning -Dyes -Confocal scanner -CCD scanner Image File Formats Image analysis -Locating the spots -Segmentation -Evaluating data quality

13 Hybridization Probe length

14 Microarray technology Scanning Images Washing

15 Labeling dyes and their properties The two most common fluorochromes used are: Cyanine3 (cy3, exicitation = 554, emission = 568) Cyanine5 (cy5, exicitation = 650, emission = 672) But Alexa dyes are also becoming popular Excitation Emission Flourescence

16 Cyanine Dye spectra excitation and emission Excitation Emission Excitation Emission

17 Alexa Dyes comparison of excitation spectra

18 Confocal scanner diagram

19 The confocal scanner scans the slide

20 CCD scanner detects from an area CCD camera Emission filter White light Beamsplitter Excitation filter

21 Microarray image formats The most common file format is 16bit TIFF. A 16bit TIFF file describe each pixel in an image with an intensity ranging from The image resolution is commonly 10mm [currently, max 5mm] Normally two scans in different wavelengths result in two monochrome files that are overlaid Pseudo-color overlay Cy 3 Channel Cy 5 Channel

22 Image analysis overview Locating the spots Segmentation Density Ensuring good data quality (flagging) Intensity

23 Locating the spot features A grid is laid over the image to aid the program in identifying the individual spots Most programs have some automation in this step.

24 Spot Segmentation Overview Fixed circle Adaptive circle Adaptive shape Histogram method ScanAlyze, GenePix, QuantArray GenePix, Dapple Spot, (a R package) ImaGene, QuantArray DeArray Density Intensity

25 Fixed Circle segmentation Fits a circle with a constant diameter to all spots in the image Easy to implement The spots need to be of the same shape and size

26 Adaptive Circle segmentation The circle diameter is estimated separately for each spot Problematic if spot exhibits non-circular shapes

27 Adaptive Shape segmentation Starts by Specifying a starting points (given by the gridding) Regions grow outwards from the starting point according to the difference between a pixel s value and the running mean of values in an adjoining region.

28 Histogram segmentation Uses a target mask chosen to be larger than any spot Foreground and background intensity are determined from the histogram of pixel values for pixels within the masked area Example : Background : mean between 5th and 20th percentile Foreground : mean between 80th and 95th percentile Density Intensity

29 Background Estimation examples Spot ScanAlyze QuantArray

30 Spot Irregularity examples Some spots may be more uncertain than others. This can be caused by: -Dust grains -Background smear -Strange shaped spot (comet tails, etc.) -Donuts shaped spots -Weak signals

31 Quality Measures examples We can pickup most of these irregularities by these measures - Intensity variability measures - Spot size deviation - Circularity deviation - Relative signal to background intensity - Position deviates from a rectangular grid Based on such measurements, a spot can be rejected

32 Feature Intensity calculation The average or median pixel value in the spot and background masks are calculated.

33 Output Example (ImaGene) Field Meta Row Meta Column Row Column Gene_ID Flag Signal Mean Background Mean A ZY A ZY A ZY A ZY A ZY A ZY A ZY A ZY A ZY A ZY A ZY A ZY A ZY

34 Examples of Difficult Images microarray images

35 Book about hybridization Nucleic Acid Hybridization (Introduction to Biotechniques S.) By: M.L.M. Anderson Paperback 256 pages (December 1998) Publisher: BIOS Scientific Publishers ISBN: