diff options
| -rw-r--r-- | .guix/hsmice-test.scm | 4 | ||||
| -rw-r--r-- | README.md | 15 | ||||
| -rw-r--r-- | e2e-tests/hsmice/check-qtl.py | 2 |
3 files changed, 12 insertions, 9 deletions
diff --git a/.guix/hsmice-test.scm b/.guix/hsmice-test.scm index ee8ecbd..9137dc6 100644 --- a/.guix/hsmice-test.scm +++ b/.guix/hsmice-test.scm @@ -49,7 +49,7 @@ (define-public r-mixed-model-gwas (package (name "r-mixed-model-gwas") - (version "1.3") + (version "1.3.1") (source (origin (method git-fetch) (uri (git-reference @@ -58,7 +58,7 @@ (file-name (git-file-name name version)) (sha256 (base32 - "0yv86mw9m981vzl80j100lg05kc6jm5ijhq9b8zcd8f2lr3115db")))) + "0vll55v8wjc0179n5q9ch9ah3dvgymc374wlbz33yzyi35yr8ds2")))) (build-system r-build-system) (home-page "https://github.com/encryption4genetics/mixed-model-gwas") (synopsis "R mixed model GWAS") diff --git a/README.md b/README.md index 2b3ab5d..de93fd2 100644 --- a/README.md +++ b/README.md @@ -63,11 +63,11 @@ pyhegp --help  -In this simple scenario, there is only one data owner and they wish to share their encrypted data with a researcher. The data owner encrypts their data with: +In this simple scenario, there is only one data owner and they wish to share their encrypted data with a researcher. The data owner encrypts their genotype and phenotype data with: ``` -pyhegp encrypt genotype.tsv +pyhegp encrypt genotype.tsv phenotype.tsv ``` -They then send the encrypted data `genotype.tsv.hegp` to the researcher. Note that data sharing is carried out-of-band and is outside the scope of `pyhegp`. +They then send the encrypted `genotype.tsv.hegp` and `phenotype.tsv.hegp` to the researcher. Note that data sharing is carried out-of-band and is outside the scope of `pyhegp`. ## Joint/federated analysis with many data owners @@ -77,17 +77,18 @@ Data owners generate summary statistics for their data. ``` pyhegp summary genotype.tsv -o summary ``` -They share this with the data broker who pools it to compute the summary statistics of the complete dataset. +They share this with the data broker who pools it to compute the summary statistics of the complete dataset. Any SNPs not common to all summaries will be dropped. ``` pyhegp pool -o complete-summary summary1 summary2 ... ``` -The data broker shares these summary statistics with the data owners. The data owners standardize their data using these summary statistics, and encrypt their data using a random key. +The data broker shares these summary statistics with the data owners. The data owners standardize their data using these summary statistics, and encrypt their genotype and phenotype data using a random key. Any SNPs not in `complete-summary` or have a zero standard deviation are dropped. SNPs with a zero standard deviation have no discriminatory power in the analysis. ``` -pyhegp encrypt -s complete-summary genotype.tsv +pyhegp encrypt -s complete-summary genotype.tsv phenotype.tsv ``` -Finally, the data owners share the encrypted data `genotype.tsv.hegp` with the broker who concatenates it and shares it with all parties. +Finally, the data owners share the encrypted `genotype.tsv.hegp` and `phenotype.tsv.hegp` with the broker who concatenates it and shares it with all parties. ``` pyhegp cat-genotype -o complete-genotype.tsv.hegp genotype1.tsv.hegp genotype2.tsv.hegp ... +pyhegp cat-phenotype -o complete-phenotype.tsv.hegp phenotype1.tsv.hegp phenotype2.tsv.hegp ... ``` Note that all data sharing is carried out-of-band and is outside the scope of `pyhegp`. diff --git a/e2e-tests/hsmice/check-qtl.py b/e2e-tests/hsmice/check-qtl.py index feae361..6e342b1 100644 --- a/e2e-tests/hsmice/check-qtl.py +++ b/e2e-tests/hsmice/check-qtl.py @@ -23,5 +23,7 @@ import pandas as pd if __name__ == "__main__": df = pd.read_csv(sys.argv[1], sep="\t") qtl = df.query("p < 1e-10") + # Assert that the QTL is on chromosome 4. assert (qtl.chromosome == 4).all() + # Assert that the QTL is within 2 Mb of the expected position. assert ((qtl.position - 137715608).abs() < 2*10**6).all() |