Is Omim No Longer Available As Structured Data?

Question

[Dave]

One of our project used to query OMIM data as XML through NCBI's efetch utility, as described here for example:

What is the best way to interact programmatically with OMIM?

However, it seems the service has stopped functioning a few months ago. It now simply returns the following error:

Database: omim - is not supported

I can find no mention of an update to the API on NCBI's website or anywhere else. At the same time, the pages accessible directly on OMIM's website offer no link to structured data (XML or otherwise) and the downloadable file, while using some specific format to delimit fields, is still far from the flexibility of the former XML files (for example, it is impossible to retrieve metadata for each reference).

Is there currently any way to regain access to OMIM data in a structured, parsable format (XML...)?

Answer 1

[Mary]

FYI: this was just announced this morning via twitter:

https://twitter.com/#!/OmimOrg/status/196939511220015104

@OmimOrg

OMIM API is now open, see http://omim.org/api and http://omim.org/help/api #OmimOrg

[Egon Willighagen]

BTW, that is not "Open". That is free access, as in free beer. You still cannot redistribute. But the API does answer the question asked. So, still upvoted. But what OMIM needs, is to be released as CC-ND-BY-SA.

Answer 2

[Pierre Lindenbaum]

I've been waiting for this since July 2011. https://twitter.com/OmimOrg/status/90770519841980416

@yokofakun[?] @mrrizkalla and we have an web API in the wings, not quite ready to open it up yet . omim.org (@OmimOrg)

[Lars Juhl Jensen]

Clearly it's their top priority, Pierre ;-)

[Egon Willighagen]

July 20122? huh?

Answer 3

[Lars Juhl Jensen]

If you go to the NCBI webpage for OMIM, this is clearly stated:

NCBI is implementing changes to help you find current content in OMIM based on resources at NCBI, and then directing you to omim.org. Please be aware that you will leave NCBI to view OMIM records. Access to full records from NCBI (e.g. web, ftp, eutils) will no longer be supported.

Sadly it looks like access control to the full OMIM data has been made more restricted by moving it to a separate site run by Johns Hopkins University. From the looks of it you can obtain the same flatfiles as you used to be able to get from the NCBI FTP site. However, I see no sign of web services. Their help pages make no mention of XML.

[Egon Willighagen]

This answer is no longer the definate answer. This is the newer answer: http://biostars.org/post/show/19421/is-omim-no-longer-available-as-structured-data/#43857

[Lars Juhl Jensen]

Thanks for following up, Egon. Good to see that the API eventually did come online, although it is not Open (but then again, OMIM never was).

[Dave]

I had missed this part (their doc for eutils still clearly give examples to use OMIM). So the short answer is no, there is no XML access to OMIM any more.

Answer 4

[Neilfws]

This is unfortunate. To summarise for anyone interested:

Esearch returns IDs (in this case, 601228).

curl "http://eutils.ncbi.nlm.nih.gov/entrez/eutils/esearch.fcgi?db=omim&term=Colorectal&retmax=1"

Efetch returns the error:

curl http://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=omim&id=601228
# Database: omim - is not supported

Esummary returns a brief XML document:

curl http://eutils.ncbi.nlm.nih.gov/entrez/eutils/esummary.fcgi?db=omim&id=601228


  http://www.ncbi.nlm.nih.gov/entrez/query/DTD/eSummary_041029.dtd">
  <eSummaryResult>
    <DocSum>
  <Id>601228</Id>
  <Item Name="Oid" Type="String">%601228</Item>
  <Item Name="Title" Type="String">POLYPOSIS SYNDROME, HEREDITARY MIXED, 1; HMPS1</Item>
  <Item Name="AltTitles" Type="String">COLORECTAL CANCER, SUSCEPTIBILITY TO, 4, INCLUDED; CRCS4, INCLUDED</Item>
  <Item Name="Locus" Type="String">15q15.3-q22.1</Item>
    </DocSum>
</eSummaryResult>

So answer to the question is "no, it isn't."

Answer 5

[Chris Maloney]

The omim.org download page indicates that the data is copyrighted by the JHU, and they make you agree not to redistribute it if you want to download it. That might be part of the reason they don't make it available through an API. But why would they feel the need to control it like this, I wonder.

[Dave]

This copyright restriction has been true from the start. The problem about now is not just that they don't provide a programmatic way to access the data, it's that they don't even provide that data (the web version is not as complete as the old XML was.