Dr Murat Kantarcioglu muratkutdallasedu Harichandan Roy harichandanroyutdallasedu Data Security and Privacy Lab UT Dallas What is Task 1 Given casecontrol datasets Desired result perturbed datasets ID: 280046
Download Presentation The PPT/PDF document "Privacy Preserving Data Dissemination" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Privacy Preserving Data Dissemination
Dr. Murat
Kantarcioglu
(muratk@utdallas.edu)
Harichandan
Roy (
harichandan.roy@utdallas.edu)
Data Security and Privacy Lab
UT DallasSlide2
What is Task 1?
Given- case/control datasets
Desired result- perturbed datasets
Case
Control
Desired Result
snps
snps
Perturbed
snps
anonymous
Method appliedSlide3
Proposed Method: NCBI Added Noise
Main Idea:
Keep
the significant SNPs
MAFs, add noise to non-significant SNPs MAFsFor significant snipsDefine significant and non-significant snips based on p-value
using pLink (significant if p is less than 0.05)Keep MAF as it is in caseSlide4
Proposed Method: NCBI Added Noise
For non-significant snips
Take variance of MAF in case
Get
gaussian random values, noises, with mean = 0 and std = (a*Math.sqrt(var)), where a = 3
.Finally, add noise to NCBI MAFIf NCBI MAF does not exist, add noise to average MAF of case and controlOrder all perturbed MAFs as in case fileSlide5
Flow DiagramSlide6
Results
For chr2,
Used NCBI Added Noise Method
Got power ≈ 8 <20
For chr10,Used Average Case/Control MethodGot power ≈ 8 <20
Power ≈ 8 <20It means desired level of privacy is preservedResults are not always sameSlide7
Other Tries
Method Applied
Result for chr2
Result for chr10
Control-
maf for non-sig
Power > 60Power < 10
Avg-maf for non-sig
Power > 60
Power < 10NCBI-
maf for non-sigPower > 40Power < 10
Control-maf/2 for non-sig
Power < 5Power >
20Avg-maf/2 for non-sig
Power <
5
Power > 30NCBI-maf/2 for non-sigPower < 5Power >
10Min-maf for sig.
Power > 40Power < 10Max-maf for sig
Power > 40Power < 10
Gaussian noise for non-sig
Power < 15
Power <10Slide8
Questions?