To detect breakpoint by cytoscan 750 K microarray technology
1
0
Entering edit mode
4.2 years ago
raya.girish ▴ 30

Hi All I need small help I have microarray affymetrix data cytoscan 750K. How can we detect breakpoint by cytoscan 750K microarray.

Breakpoint • 1.4k views
2
Entering edit mode
4.2 years ago

For that microarray technology and for what you want to achieve, I would use the tools from the Aroma Project. In particular, you should follow the vignette for CRMA (v2): Estimation of total copy numbers using the CRMA v2 method. This consists of CRMA normalisation followed by circular binary segmentation (CBS) on the derived estimated copy number at each site:

In following the vignette, for the annotation loading section, you will want to run this for your chip type:

> cdf <- AffymetrixCdfFile$byChipType("CytoScan750K_Array") > cdf AffymetrixCdfFile: Path: annotationData/chipTypes/CytoScan750K_Array Filename: CytoScan750K_Array.cdf Filesize: 161.10MB Chip type: CytoScan750K_Array RAM: 0.00MB File format: v4 (binary; XDA) Dimension: 2166x2166 Number of cells: 4691556 Number of units: 750865 Cells per unit: 6.25 Number of QC units: 4  [source: http://www.aroma-project.org/chipTypes/CytoScan750K_Array/] ------------------------------------ Finally, this is implemented in the aroma.affymetrix R package. Kevin ADD COMMENT 0 Entering edit mode Hi Kevin But this will not show me whether there is unbalanced or balanced translocations ADD REPLY 1 Entering edit mode Okay, in any case, use the method to at least derive the copy number, and then use those values with another program to get exactly what you want. Can you elaborate on 'balanced' and 'unbalanced' translocations? - you originally only mentioned that you had wanted to identify breakpoints. ADD REPLY 0 Entering edit mode I have POC sample product of conception (Aborted child). Both parents are showing LOH and child is showing gain loss and LOH all three. What we are suspecting that death of child was due to unbalanced translocation of chromosome because have balanced translocation. So i am given the task for detection of breakpoints and unbalanced region from Cytoscan 750K in child ADD REPLY 0 Entering edit mode Sounds very interesting and important. You would have to therefore derive allele-specific copy number, and there appears to be a previous publication for that involving the CytoScan 750K: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5086940/ ADD REPLY 0 Entering edit mode What is the difference between the R package you provided earlier and this rawcopy? ADD REPLY 1 Entering edit mode They can both derive allele-specific copy number, from what I can see. However, I favour the second one that I have found because it seems easier to implement (I am not sure of your level of bioinformatics, and I am aware that Aroma can be difficult to follow), and it was also specifically designed for the high-density Affymetrix arrays (SNP 6.0 and CytoScan). ADD REPLY 0 Entering edit mode Okay Kevin , Thanks for the help. ADD REPLY 0 Entering edit mode Hi Kevin This script is not working for me cdf <- AffymetrixCdfFile$byChipType("CytoScan750K_Array")

I am unbale to find CDF file for Cytoscan750_Array in Bioconductor. If it is not available there can you please tell me where will i get that file .

1
Entering edit mode

You will have to download the CDF for the CytoScan 750K, and then place i in a directory called ./annotationData/chipTypes/CytoScan750K_Array/

The file should be saved as: CytoScan750K_Array.cdf

It seems that Aroma and the CytoScan are not well supported anymore, but I think that you can download the CDF from here (bundled with other files): https://bitbucket.org/brge/affy2sv/wiki/CytoScan

If you cannot get it working, then the other program / R package (rawcopy) seems promising.

0
Entering edit mode

Kevin Just want to confirm I am in correct path breakpoint will loss or gain at particular position of any arms right. So for example I have Loss in chr12 with position showing 12q31(10-100) X3 So loss is deletion at position 12q31 start coordinate 10 and end coordinate 100. This will be my breakpoint

0
Entering edit mode

Okay, that would be a small region, though, with just 90bp. It may be more appropriate to just refer to it as a copy number alteration (CNA). For breakpoints, we usually refer to larger section of DNA, i.e., larger genomic aberrations.

How did you derive the copy number? - rawcopy?

When we derive copy number for each probe region, we then typically agglomerate the copy number values into larger regions, such that we can identify things like this:

Those are both from my PhD (8 years ago) and using Affymetrix SNP 6.0. Back then, algorithms were not great for copy number. In the first, there is a possible breakpoint at 4q22.2. The second figure is indicative of iso-cromoome 8 (deletion of 1 arm; amplification of the other).

So, have you already derived copy number from the samples? Can you share some code? rawcopy should have some plotting or agglomeration function. You can open a new question if you want to get more fresh ideas. If you have already derived copy number, then that is a good starting point for diverse types of analyses. I was worried that the CytoScan 750K was no longer supported, so, sucess in that sense.

Kevin

0
Entering edit mode

This is the result i got from Rawcopy. https://ibb.co/c02TYn

How should i interpret it now also how will i be able to whether its unbalanced or balanced translocations

0
Entering edit mode

There does not appear to be any copy number aberrations...

0
Entering edit mode

Okay. But can i detect unbalanced translocations using Cytoscan750K microarray and using R for cytoscan 750K

0
Entering edit mode

Possibly not - why did you choose the CytoScan 750k? You can certainly identify breakpoints with CytoScan 750K, which was your original question.

To detect de novo translocations, you would have to do whole genome sequencing and preferably phased whole genome sequencing, which is very expensive.

If not de novo identification, I presume that you have a target region of interest?

0
Entering edit mode

no with this breakpoint i need to make probes for fish. But my customer is saying to me that it is possible with microarray to check for unbalanced translocations.

0
Entering edit mode

I don't recall seeing translocation information from microarray, but I do not doubt that it is technically feasible.

Have you come across the 'CytoScan Cytogenetics / Optima Suite'? - take a look Here. In addition, you may be interested in Chromosome Analysis Suite.

Apologies for sending you on wild goose chases, but the CytoScan never really took off and is not supported that much, as I mentioned. I attended a meeting with the Affy CEO a few years ago, in fact, and have his business card here.