Bulk Data

Bulk data can be accessed at the iCite database snapshot repository.

API

/api/pubs

List publications, including data about each one.

Optional query parameters:

• limit: how many publications to return. The maximum allowed is 1000.
• offset: only return publications with a PMID greater than this.
• year: only return publications from the given year.
• pmids: only return publications with the given PubMed IDs. Separate multiple IDs with commas to request up to 1000 at a time. If this parameter is provided, all other parameters are ignored.
• fl: only return publications with the given fields. Separate multiple fields with commas (no space). Field names are very specific and listed in Response example below. No fl param will return all fields.
• format: return data as a csv (comma separated value) by specifying format=csv rather than the default JSON.
  Example: /api/pubs?offset=23456789&limit=10&format=csv
Example: /api/pubs?pmids=28968381,28324054,23843509&fl=pmid,year,title,apt,relative_citation_ratio,cited_by_clin

/api/pubs/{pmid}

Get data on a specific publication. Example: /api/pubs/23456789

/api/pubs?pmids={pmid1,pmid2...}

Get data on multiple specific publications.
Example: /api/pubs?pmids=23456789,27599104
Example: /api/pubs?pmids=23456789,27599104

Code Examples

Get the data for a specific PMID by visiting /api/pubs/23456789:

{
    pmid:	                23456789
    year:	                2013
    title:	                "Hospital volume is associated with survival but not multimodality therapy in Medicare patients with advanced head and neck cancer."
    authors:                    "Arun Sharma, Stephen M Schwartz, Eduardo Méndez"
    journal:	                "Cancer"
    is_research_article:	"Yes"
    relative_citation_ratio:    1.73
    nih_percentile:	        70.3
    human:	                1
    animal: 	                0
    molecular_cellular:	        0
    apt:	                0.75
    is_clinical:	        "No"
    citation_count:	        26
    citations_per_year:	        4.33333333333333
    expected_citations_per_year:2.50953735885351
    field_citation_rate:	4.87055203724375
    provisional:	        "No"
    x_coord:	                0
    y_coord:	                1
    cited_by_clin:              []
    cited_by:	                […]
    references:	                […]
    doi:	                "10.1002/cncr.27976"
    last_modified:	        "05/15/2022, 01:32:57"
}

# If iCiteR is not installed,
# install it with: install.packages("iCiteR")

library(iCiteR)
papers = get_metrics(c(23456789,27599104))
print(papers)








        

import requests


response = requests.get(
    "/".join([
        "https://icite.od.nih.gov/api",
        "pubs",
        "23456789",
    ]),
)
pub = response.json()
print(pub)


        

var request = new XMLHttpRequest();
request.open(
    "GET",
    [
        "https://icite.od.nih.gov/api",
        "pubs",
        "23456789",
    ].join("/")
);
request.responseType = "json";
request.onload = function () {
    console.log(request.response);
};
request.send();
        

Content Type

By default, responses are JSON. CSV responses are supported as well and can be requested via Accept: text/csv.

Search Link

If user wants to generate search link programmatically, the Python code below demonstrates one way to do that.

Code Example

            import requests
            import json
            def create_link(pmid_str):
                base_url = "https://icite.od.nih.gov"
                result_url_base = "https://icite.od.nih.gov/results"
                try:
                    # Step 1: Submit PMIDs to get a search ID
                    headers = {'Content-Type': 'application/json'}
                    payload = {'pmidText': pmid_str}
                    response = requests.post(base_url + "/searchid", headers=headers, json=payload)
                    response.raise_for_status()  # Raise an error for HTTP status codes 4xx/5xx
                    data = response.json()
                    search_id = data.get('search_id')
                    if not search_id:
                        raise ValueError('Search ID not found in response')
                    # Step 2: Use the search ID to get results
                    result_url = f"{result_url_base}?search_id={search_id}"

                    return result_url

                except requests.exceptions.RequestException as e:
                    print(f"Request error: {e}")
                    raise
        

Feedback

Please send any feedback to iCite@mail.nih.gov.