PrivMAF is a test that can be used on minor allele frequency data to determine how much releasing the minor allele frequencies from the study will compromise the privacy of those who participated in the study. You can download our implementation of PrivMAF by clicking here. You can also download the readme file with instructions on use by clicking README. Derivations of our results can be downloaded here.
In order to run PrivMAF.py a user needs to have python installed, as well as numpy. Beyond that there are no other dependencies. For details on how to run the program see the README file.
We also include an example composed of fake data. The data files are fakeData.tped and fakeMAF.txt. Running this with N=10000 we get an PrivMAF score of .0463. This corresponds to the command:
python privMAF.py fakeData.tped fakeMAF.txt 100000
We have also implemented our release mechanism, ALGT. Details are given in the README file.
Any questions or concerns can be directed to seanken at mit dot edu