Example query to fetch BXD_10007:

SELECT
    px.Id,
    s.Name,
    pd.Value,
    pse.error
FROM PublishData pd
INNER JOIN Strain s
    ON pd.StrainId = s.Id
INNER JOIN PublishXRef px
    ON pd.Id = px.DataId
INNER JOIN PublishFreeze pf
    ON px.InbredSetId = pf.InbredSetId
LEFT JOIN PublishSE pse
    ON pse.DataId = pd.Id
   AND pse.StrainId = pd.StrainId
LEFT JOIN NStrain ns
    ON ns.DataId = pd.Id
   AND ns.StrainId = pd.StrainId
WHERE
    pf.Name = 'BXDPublish'
    AND px.Id = 10007
    AND pf.Public > 0
    AND pd.Value IS NOT NULL
    AND pf.Confidentiality < 1
ORDER BY
    LENGTH(s.Name),
    s.Name;

• [x] (alexm) Debug genenetwork3 localsetup
• [x] (alexm) script to dump all phenotypes to lmdb see pr

• https://github.com/genenetwork/genenetwork3/pull/240

• [x] (alexm) RESTful api URI: "/dataset/phenotype/<str:dataset>_<integer:trait_id>" e.g: "https://cd.genenetwork.org/lmdb/v1/data/traits/BXDPublish_10007". Result set should have values, SE.

Comments: PR in genenetwork3

• https://github.com/genenetwork/genenetwork3/pull/240

• [X] Deploy functionality to tux02

Prototype:

• https://github.com/Alexanderlacuna/gn_lmdb_rdf_interface/tree/master

• [X] (bonfacem) Modify endpoint to use json extension:

curl http://127.0.0.1:8091/dataset/bxd-publish/values/10002.json

• [X] (alexm) Remove null values from data end-point.
• [ ] (bonfacem, pjotrp, alexm) How to work with case-attributes metadata.
• [ ] (bonfacem) Add above link to RDF.

• [ ] Case Attributes.
• [ ] Individual Strains.
• [ ] (GEMMA/ Rqtl) Compute data.

• [X] (bonfacem) Create a directory for Alex to work against.
• [ ] (alexm) Create an lmdb view with a dataset that has the largest N-row x N-columns view. See output. Small experiment with correlations.
• [ ] Mount LMDB_DATA_PATH to another partition because of "size" constraints(?)

TODO

• [ ] Add g-exp scripts for generating all the data.
• [ ] Figure out how to add geno-files (or the lmdb view) to RDF.
• [ ] Replace naming scheme to use new GN RDF naming convention. See:

• https://issues.genenetwork.org/issues/rdf/rdf-refinement

Old scheme, combine trait_id (PublishXRef) + Dataset Name (PublishFreeze.Name). e.g.

• https://genenetwork.org/show_trait?trait_id=12315&dataset=BXDPublish

New scheme should mimic:

• https://rdf.genenetwork.org/v1/id/trait_BXDPublish_IOP_all_ages