Pubchem Patents Bulk Download

Question

[nmegz4ever]

The ftp download site of pubchem gives only downloads for Compound information. Is there a way to bulk download patent data for each compound?

[ATpoint]

Please provide some details what exactly you need including a link to the respective page. See How To Ask Good Questions On Technical And Scientific Forums

Answer 1

[Jeremy Leipzig]

here is some python code you can use if you have a CID in mind

cid=2446
pubchemapi = "https://pubchem.ncbi.nlm.nih.gov/rest/pug/compound/cid/{0}/xrefs/PatentID/JSON".format(cid)
try:
    url=urllib.request.urlopen(pubchemapi)
except urllib.error.HTTPError as err:
    print("tried {} will sleep on it".format(pubchemapi))
    time.sleep(5)
    try:
        url=urllib.request.urlopen(pubchemapi)
    except urllib.error.HTTPError as err:
        print("can't find {0}".format(cid))
        return(None)
pbcresp = json.loads(url.read().decode())
patents = pbcresp['InformationList']['Information'][0]['PatentID']

for patent in patents:
    if patent.startswith('US'):
        patentapi = "https://pubchem.ncbi.nlm.nih.gov/rest/pug_view/data/patent/{0}/JSON?heading=Patent+Submission+Date".format(patent)
        try:
            url=urllib.request.urlopen(patentapi)
        except urllib.error.HTTPError as err:
            print("tried {} will sleep on it".format(patentapi))
            time.sleep(5)
            try:
                url=urllib.request.urlopen(patentapi)
            except urllib.error.HTTPError as err:
                print("can't find {0}".format(cid))
                continue
        patresp = json.loads(url.read().decode('latin-1'))
        submissiondate = patresp['Record']['Section'][0]["Information"][0]["Value"]["DateISO8601"][0]
        print("{0} {1} {2}".format(cid,patent,submissiondate))