Technical issues and case interpretations PGD and PGS User Group Meeting April 8th, 2013 Zuzanna Bieniawska
© 2013 BlueGnome. All rights reserved. .
Overview
This presentation summarises most common technical issues during array processing and toughest cases interpretation
2
►
Good and bad reference dataset
►
Good and low X separation
►
Different triploidy types and what BlueFuse can detect
►
Resolution of 24sure+ array
►
Software problems
Case 1 – importance of good reference dataset
“Can you please explain discrepancy in these 2 cases?” One of our customers run the same amplified trophectoderm DNA twice in 2 different DNA batches: ►
Experiment 1
labelling 15h, hybridization 5h ►
Experiment 2
labelling 5h, hybridization 16h
Customer decided to repeat the experiment because initially the saw low signal on the array
3
Case 1 – importance of good reference dataset
Experiment 1, -16, XY labelling 15h, hybridization 5h
4
Experiment 2, -16, -19, XY labelling 5h, hybridization 16h
Case 1 – importance of good reference dataset
female vs female
male vs male
Experiment 1, -16, XY labelling 15h, hybridization 5h
5
Experiment 2, -16, -19, XY labelling 5h, hybridization 16h
Case 1 – importance of good reference dataset - solution
6
►
There is big difference in how these 2 experiments were performed but both experiments were performed according to protocol guidance so it should not be the case for discrepancy in results.
►
Therefore we think it is because performance in 2 reference sets were so different in both experiments and this influenced performance of the samples.
►
Our guidelines say that % of included clones should be around 90% or more and this is especially important for references as performance of each sample is calculated based on references .
►
In this case it was much lower than what we recommend: 65% and 72% this is something we would probably internally fail if we were running protocol here.
Importance of good separation of sex chromosomes
Good X/Y separation ►
X-separation does vary with hybridization time, ranging from around 0.4 for a 3hr hybridization to 0.9 for a high quality overnight hybridization.
►
X-separation should be consistent across samples in experimental batch
Technical reasons for low X separation ►
Incorrect temperature of wash 3 (usually too low) - always check the temperature of the buffer with external thermometer. Please also always allow at least 40min for 0.1SSC buffer to warm up to 60C. Rewashing of arrays may work (although we can not guarantee there is any signal left on the array).
►
Not sufficient amounts of COT-1 added during combination step.
Biological reasons for low X separation ►
7
Abnormalities of sex chromosomes
Importance of good separation of sex chromosomes
Bad X/Y separation; X separation of 0.2 indicates a failed experiment.
8
Good X/Y separation
Case 2 – 68 XXY, -14 triploid case – blastomere Can someone give us second opinion about this case?
9
Case 2 - triploid XXY case - blastomere
10
Case 2 – 69 XXY triploid case - blastomere
In this example the software has established greater similarity with the male reference 11
Case 2 – 68 XXY,-14 triploid case – blastomere
In this example the software has established greater similarity with the female reference, so sex chromosomes are opposite to the example 12
Case 3 - triploid XXX case - blastomere
►
Recent PGD case for hemophilia gene; 2 embryos screened by PCR for mutation in hemophilia gene, STR analysis indicated that one of the embryo resulted in XXX
►
Both embryos were screened on 24sure for aneuploides. One embryo was euploid one anueuploid but both embryos showed the same pattern in terms of sex chromosomes separation
►
How to explain it?
13
Case 3 - triploid XXX case - blastomere
Unfortunately BlueFuse Multi is unable to differentiate an XX and an XXX on 2 which is showing 4 copies.
14
Case 4 – can 24sure+ detect….
►
►
►
15
“I am contacting you with regards to one of our recent PGD requests, which involves a 2.5Mb duplicated region on chromosome 22 (q11.21). We were thinking if is possible to detect this duplication with 24sure+ arrays.
“Could you please confirm my suspicions that a 263kb Xq28 duplication is too small to be detected with 24sure+?”
We have a case of 18q duplication: 46,XX,dup(18)(q21.33q23). Is it possible to do PGD with 24sureV3 or 24sure+ for this duplication
Case 4 – can 24sure+ detect….
►
24sure + is the platform for detection of unbalanced translocations (so we see result on both chromosomes) and not for known microdeletion/duplication syndromes
►
It can detect ~2MB (at minimum) imbalances ~7 BACs
►
If you are unsure if the array can detect the duplication/deletion you can “do a workup for the case”.
Count/look how many BACs are in the region of interest You would extract genomic DNA from the carrier parent, dilute it down to 15pg and run that sample through the amplification process. The amplified DNA can then follow through the 24sure+ system and you would be able to tell confidently if the duplication can be observed.
16
Case 4 – can 24sure+ detect….
►
“I am contacting you with regards to one of our recent PGD requests, which involves a 2.5Mb duplicated region on chromosome 22 (q11.21). We were thinking if is possible to detect this duplication with 24sure+ arrays. NO, you will be better of with STR analysis
►
“Could you please confirm my suspicions that a 263kb Xq28 duplication is too small to be detected with 24sure+?” NO, the region is too small
►
We have a case of 18q duplication: 46,XX,dup(18)(q21.33q23). Is it possible to do PGD with 24sureV3 or 24sure+ for this duplication Probably it is 13MB change, 42 BACs in total
17
Case 5 – BlueFuse
What is the Annotation Database? ►
The information included in the annotation DB is used by BlueFuse Multi during analysis of the different array types (influences correct algorithm settings per array/experiment). The information displayed in the DecisionTrack pane is also derived from the annotation DB files.
►
Annotation databases have been updated and released due to reflect changes in the genomic information and associated data now available in the public domain or other licensed data sources.
►
When you set up new database copy the latest annotation DB locally to your PC, and from there import it to BlueFuse Multi (tools->import new annotation DB)
►
Correct size ~600MB
18
Case 5 – BlueFuse – corruption of annotation db
If you store your databases on the network (server) and you have BlueFuse Multi 3.0 or above it may happen you will get this…
►
“I can not display my decision tracks”
►
“I was importing my data and BlueFuse stopped working suddenly”
►
“I am getting error message can not upgrade product type”
If this happens the most likely scenario is the corruption of one of annotation databases. Especially if more than one computer have BlueFuse Multi installed but only one encounters problems.
19
Case 5 – BlueFuse - corruption of annotation db
►
The normal (and necessary) behaviour is when you run BlueFuse Multi over network all annotation files will be copied locally when you start a database.
►
If you delete these files, as soon as you open BlueFuse Multi it will recopy the files from the main database back into this folder otherwise the software will not work
►
You can find this files….
Windows XP: C:\documents and settings\all users\application data\bluegnome\bluefuse multi
Vista and Windows 7: C:\program data\bluegnome\bluefuse multi
20
Case 5 – BlueFuse - corruption of annotation db - solution
►
Close down BlueFuse Multi
►
Delete corrupted annotation db files
►
Open BlueFuse Multi, login and it will copy these annotation DBs back into the folder.
►
Check that annotation db files have come across, and they are the correct size in Kb. If they are too small it is likely local disk is full and cannot copy entire file, this will not throw an obvious error or log.
►
In most cases this will fix all problems, if annotation looks correct, but if there are still issues, probably a reinstall is only solution
21
Case 5 – BlueFuse – corruption of annotation db
If you store your data locally on PC it may happen you will get this as well…
►
“I can not display my decision tracks”
►
“I was importing my data and BlueFuse stopped working suddenly”
If this happens the most likely scenario is the corruption of one of annotation databases that is stored in GAL and CLONE folder in your database
22
Case 5 – BlueFuse - corruption of annotation db - solution
►
Close down BlueFuse Multi
►
Copy the correct annotation db from our webpage to your PC
►
Delete corrupted annotation db files from GAL and CLONE folder (you may need to first unhide all folders) and copy the right one to the folder (or preform copy and replace with the right one)
►
Check that annotation db files have come across, and they are the correct size in Kb. If they are too small it is likely local disk is full and cannot copy entire file, this will not throw an obvious error or log.
►
In most cases this will fix all problems, if annotation looks correct, but if there are still issues, probably a reinstall is only solution
23
Case 6 – BlueFuse Multi – lack of free space
“I was importing my data and BlueFuse stopped working suddenly” “My BlueFuse was fine yesterday but it crashed today”
BlueFuse Multi is not giving you any other obvious error or extra information
Please check free space where you keep your databases
BlueFuse Multi may refuse working even if there are still 10GB available
24
What sort of problems do you have when running 24sure/24sure+ protocol ►
Do you get false positive/negative call on Y chromomose
►
Do you get suppressed X/Y separation ratio
25