Variant to Gene Query for multiple variants in OT Genetics

Data Access GraphQL API
1

Dear Open Target team,

First of all, thank you very much for your valuable work.

I need to perform a Variant to Gene search for multiple variants (>200).

I am trying to find a way to do it through a script that generates a table with the results containing the same parameters as when I search the rsid individually through the browser but I’ve been not able to find what I am looking for in the documentation.

Your guidance would be very appreciated.

Thank you very much!

Paloma

2

Thanks for your question Paloma. For a moderate number of variants, you could use the GraphQL API, which can be access via a browser, or programmatically.

For example, a query in the browser might look something like this:
GraphQL variant to gene query.

You can use a language of your choice to do the query programmatically, e.g. python or R. Here is an example in R:

# Install relevant library for HTTP requests
library(httr)

# Set gene_id variable
variantId <- "17_44352876_C_T"

# Build query string
query_string = "
query v2g($variantId: String!) {
  genesForVariant(variantId: $variantId) {
    gene {
      id
    }
    variant
    overallScore
    distances {
      sourceId
      aggregatedScore
      tissues {
      	distance
      }
    }
  }
}"

# Set base URL of GraphQL API endpoint
base_url <- "https://api.genetics.opentargets.org/graphql"

# Set variables object of arguments to be passed to endpoint
variables <- list("variantId" = variantId)

# Construct POST request body object with query string and variables
post_body <- list(query = query_string, variables = variables)

# Perform POST request
r <- POST(url=base_url, body=post_body, encode='json')

df = content(r)
# Print first entry of V2G data console
head(content(r)$data$genesForVariant, 1)

# Flatten the nested result fields into a dataframe
library(rlist)
list_result = content(r)$data$genesForVariant
x = lapply(list_result, list.flatten)

library(dplyr)
df = bind_rows(x)

Note that if you want to get the information for individual QTL associations, you would need to do a bit more to flatten the resulting nested lists.

1 Like
3

Thank you very much for your detailed and very helpful response Jeremy! It worked!

1 Like
4

Hi JeremyS, thank you for your valuable information!

How should I change “query_string” to get the data shown as the following?

image
image1191×369 36.1 KB

I tried with the full text shown in your example, however, I got a lot of qtls, I did not see anything such as PCHi-C, DHS-promoter corr, VEP… Maybe I missed it.

I tried to find the information on the data structure and failed. Maybe I missed it too.

5

Hi llg, in the v2g_scored file, you will find this information, each row represents an evidence source for a given variant to gene pair (as indicated by the type_id and source_id columns), so if you were particularly interested in pchic evidence sources, you can filter the v2g_scored file based on the type_id column.

Hope this helps,
Xiangyu

root
 |-- chr_id: string (nullable = true)
 |-- position: long (nullable = true)
 |-- ref_allele: string (nullable = true)
 |-- alt_allele: string (nullable = true)
 |-- gene_id: string (nullable = true)
 |-- feature: string (nullable = true)
 |-- type_id: string (nullable = true)
 |-- source_id: string (nullable = true)
 |-- fpred_labels: array (nullable = true)
 |    |-- element: string (containsNull = true)
 |-- fpred_scores: array (nullable = true)
 |    |-- element: double (containsNull = true)
 |-- fpred_max_label: string (nullable = true)
 |-- fpred_max_score: double (nullable = true)
 |-- qtl_beta: double (nullable = true)
 |-- qtl_se: double (nullable = true)
 |-- qtl_pval: double (nullable = true)
 |-- qtl_score: double (nullable = true)
 |-- interval_score: double (nullable = true)
 |-- qtl_score_q: double (nullable = true)
 |-- interval_score_q: double (nullable = true)
 |-- d: long (nullable = true)
 |-- distance_score: double (nullable = true)
 |-- distance_score_q: double (nullable = true)
 |-- overall_score: double (nullable = true)
 |-- source_list: array (nullable = true)
 |    |-- element: string (containsNull = true)
 |-- source_score_list: array (nullable = true)
 |    |-- element: double (containsNull = true)
1 Like