LAMHDI Partner Starter Kit

Welcome to LAMHDI, the initiative to link animal models to human disease.  LAMHDI is designed to accelerate the research process by providing biomedical researchers with a simple, comprehensive Web-based resource to find the best animal model for their research.  For researchers, the search for animal models begins here.

The LAMHDI website provides researchers access to two types of information.  One is a structured database containing information about animal models available for human disease research, including links to further information and details on how to access particular animal models for research purposes.  The other is a selected collection of Web pages on animal models in the context of human disease research.

This technical document is for use by institutions or individual researchers curating or publishing information of interest to the LAMHDI community, and particularly institutions and researchers who make animal models available for others to use in human disease research.

Using the technical information provided in this document, partners can coordinate with the LAMHDI project team to ensure that their animal models and Web pages are appropriately indexed and are listed correctly when visitors carry out searches using LAMHDI.

How do I get my animal models into the LAMHDI index?

LAMHDI offers a search of "structured resources" representing individual animal models available to human disease researchers.  This is a database that enables researchers to locate animal models that might be relevant to their particular research programs by specifying species, disease, genotype, phenotype, or other search keywords.

Whether you have one animal model or many, if information about your model or models is in some kind of structured format (spreadsheet, database, well-ordered lab book), we invite you to become a LAMHDI structured resource partner.  As such, you determine: 

• which models to share through LAMHDI;
• how each model record is indexed and what keywords will "hit" the model;
• how it will appear on LAMHDI search result pages: what information will be displayed in each column for each model.

Depending on the number of models you offer, the depth and structure of the information describing your models, and on your technical and operational capacity to reformat and share your information, we offer several methods for getting your models into the LAMHDI index. These methods range from a one-off import (with occasional updates should you wish to refresh your data) to a fully automated bidirectional web-services interface between our two databases.

For each model, we display five columns on LAMHDI search result pages, but these columns mask broader underlying searchable information about each model.  For example, though we might only display "Mouse" or "Mus Musculus" in the Species column, a visitor's search on either term should find the same records.  We can achieve this by synonym expansion at search time, and also by indexing multiple alternative terms at index time if that is easier for you or works better for your dataset.

The LAMHDI structured resource database includes the following attributes for each animal model record:


We accept submissions in numerous formats via several transmission channels.  Acceptable formats include flat text files (CSV, TSV, pipe-delimited, et cetera), with columns ordered as listed in the table above excluding the last five attributes; XML files (schemas available at http://www.lamhdi.org/xsd/animal_models.xsd and http://www.lamhdi.org/xsd/animal_model.xsd); and JSON files (using the naming conventions codified in the XML schemas).

Transmission channels include one-off e-mail; FTP/SFTP or HTTP/HTTPS drop-offs; scheduled periodic FTP/SFTP or HTTP/HTTPS pick-ups; and a REST-like LAMHDI Maintenance API (now under construction; please contact us for details or to contribute).

As a partner, you may elect to provide LAMHDI a complete dataset each time an update is required, in which case you need not supply LAMHDI Model IDs or even Partner Model IDs.  If you wish to provide incremental updates, you must provide the LAMHDI Model ID, the Partner Model ID, or both, for each record to be updated.

While you have great latitude in the HTML markup you provide to LAMHDI to be shown when your records are displayed on LAMHDI search result pages, we require:

• that your HTML be fully valid XHTML 1.0 markup;
• that you maintain and ensure the availability of any images or other embedded content your markup references;
• that you not include or embed any executable content (Javascript, Flash, et cetera) in your markup;
• that you agree LAMHDI has the right and responsibility to strip any potentially insecure material from your markup when it is displayed on LAMHDI search result pages.

We also require that you protect the privacy of LAMHDI users to at least the same degree as LAMHDI itself protects that privacy.  In particular, you should not set persistent cookies when serving images or other external resources referenced by your markup.

As a LAMHDI structured resource partner, you will be granted access to an administrative area of the LAMHDI website where you and your designated colleagues and staff may manage the basic partner information described in the next section, as well as administrative records relating to your structured resources (format details, transmission details and schedules, et cetera).  LAMHDI will monitor all activity, ensure that only people who have appropriate approvals may make certain kinds of changes, and keep an audit trail.  You will also be assigned a LAMHDI contact person and you will have access to other support resources.

To get started, please e-mail lamhdi-info@lamhdi.org, briefly describing your interest and the general nature of the dataset you wish to share through LAMHDI.

How do I get my Web pages into the LAMHDI index?

LAMHDI employs a Web search engine, similar to Google and other search services, focused on just that portion of the Web concerned with animal models in the context of human disease research.  Our software crawls the Web, indexing this material.  We call these indexed pages "Web resources," as opposed to the "structured resources" described earlier.

If you publish material that fits within LAMHDI's mandate, we would love to include it in our index.  Once we confirm the suitability of your material, we'll need a number of details:

• basic information about your institution and your site (or sites), which you will enter and maintain using your upgraded LAMHDI user account after approval, including:
• the name of your institution (and an optional short form),
• contact details for (or the LAMHDI usernames of) any additional individuals who can discuss your material and your site (ideally, both an animal model expert and someone familiar with the technical characteristics of your site),
• contact details for (or the LAMHDI usernames of) any additional individuals who should also be authorized to maintain your LAMHDI partner information,
• the name of your site (and an optional short form),
• the "seed" URLs where you suggest we begin our crawl of your site,
• the maximum depth to which we should crawl each given seed to include all relevant material yet avoid indexing material outside the LAMHDI mandate,
• information about how often your material changes and how often you think we should re-crawl your site to update the LAMHDI index;
• your permission to have our Web indexing robot crawl your site, which may involve customizing any robots.txt file currently in place to block other search engines, or your authorization to ignore your robots.txt file when we crawl your site;
• a way for our robot to reach each page of interest on your site by following a standard hypertext link (as opposed to filling in and submitting a search form, executing a Javascript function, parsing a PDF file, or any other non-standard mechanism).
  

As a LAMHDI Web resource partner, you will be granted access to an administrative area of the LAMHDI website where you and your designated colleagues and staff may enter and manage the basic partner information and crawl seeds information described above.  As we do with our structured resource partners, we will monitor all activity, ensure that only people who have appropriate approvals may make certain kinds of changes, and keep an audit trail.  You will also be assigned a LAMHDI contact person and you will have access to other support resources.

To get started, please e-mail lamhdi-info@lamhdi.org, describing your interest and providing links to some of the material on your site you believe fits the LAMHDI mandate.

How long will it take to see my data in LAMHDI?

We strive to vet and approve new partner applications as quickly as possible, typically within two business days after receipt.

After approval, we offer immediate assistance to help new LAMHDI Web resource partners complete their basic information profiles and set up their initial crawl seeds right away.  Once the seeds are in place, our nightly crawl will automatically begin indexing the new seeds, so results can begin appearing in LAMHDI search results as soon as the following day.

LAMHDI structured resource partners may face somewhat longer initial setup times because of the work typically necessary to put structured data into a format LAMHDI can accept.  However, we also offer immediate assistance to these partners as well.  Once the initial setup and testing are complete, automated batch submissions are handled as often as nightly (or as often as you make a new dataset available).  Incremental additions, updates, and deletions made using the LAMHDI Maintenance API accumulate and are processed nightly as well, so changes made one day will be reflected in LAMHDI search results the following day.

Where to go from here?

Please e-mail lamhdi-info@lamhdi.org for further information.