MR causal estimate

Pre-computed Mendelian randomization results


Exposure trait


P-value threshold: 1e-5
Outcome trait


P-value threshold: 1e-5
Exposure trait
Outcome trait


P-value threshold: 1e-5




This view provides automated causal estimates between pairs of traits using the results from MR-EvE (Hemani et al, bioRxiv 10.1101/173682). Note that these are preliminary results generated using an automated algorithm, so should always be validated by performing an analysis with MR-Base.

Three views are available:

  • Searching for anything that an exposure might cause
  • Searching for anything that might cause an outcome
  • Retrieving the causal estimate between a single exposure and outcome

The Query tab shows how this query works and provides the API call to replicate this programmatically.

Data sources

  • MR-EvE (Hemani et al, Automating Mendelian randomization through machine learning to construct a putative causal map of the human phenome, bioRxiv 10.1101/173682)

How to use

All three approaches comprise a search form in which you enter trait values and select a threshold. After clicking search you can view the top results in the Network plot tab or the full set of results in the Table tab. The Query tab provides the information you need to download the data using the API.

The three views are accessed by pressing one of the three buttons: Search exposure trait, Search outcome trait or Search exposure and outcome traits.

Once you have selected a view, enter the trait name(s), select a p-value threshold, then click Search.

View the top results in the Network plot tab, access and search the full results in the Table tab, or use the Query tab to find out how to download data using the API.

Changing the threshold: If you move the slider you will need to click Search again to apply the new threshold


The following examples illustrate the three types of search:

 Displays 6 / 6 paths


outcome_id outcome_name exposure_id exposure_name estimate se p ci_low ci_upp method selection moescore
7 Coronary heart disease 974 Body mass index 0.388520 0.049338 3.417302e-15 0.279017 0.498022 FE IVW DF 0.89
7 Coronary heart disease 2 Body mass index 0.397101 0.072745 4.793827e-08 0.254523 0.539680 Simple median HF 0.92
7 Coronary heart disease 95 Body mass index 0.454937 0.093067 1.017140e-06 0.272530 0.637344 Penalised median DF 0.78
8 Coronary heart disease 2 Body mass index 0.330890 0.068387 1.308513e-06 0.189942 0.471837 FE IVW DF 0.75
7 Coronary heart disease 835 Body mass index 0.359686 0.075569 1.938765e-06 0.211573 0.507798 Simple median Tophits 0.91
8 Coronary heart disease 974 Body mass index 0.328476 0.073081 6.966327e-06 0.161582 0.495370 FE IVW DF 0.85
MATCH (t1:Gwas)-[r:MR]->(t2:Gwas) WHERE t1.trait = "Body mass index" AND t2.trait = "Coronary heart disease" AND r.p < 1e-05 RETURN AS exposure_id, t1.trait AS exposure_name, AS outcome_id, t2.trait AS outcome_name, r.estimate AS estimate, AS se, r.p AS p, r.ci_upp as ci_upp, r.ci_low AS ci_low, r.selection AS selection, r.method AS method, r.moescore AS moescore ORDER BY r.p ;
curl -X 'GET' -H 'Accept: */*' -H 'Accept-Encoding: gzip, deflate' -H 'Connection: keep-alive' -H 'User-Agent: python-requests/2.22.0' 'http://api.epigraphdb.org_private/mr?exposure=Body+mass+index&outcome=Coronary+heart+disease&pval_threshold=1e-05'