Surrounding a phrase within double quotes you can search for a specific phrase that spans multiple words. e.g. to search for breast cancer you can use: "breast cancer" . This will ensure that your results match your query as intended as adjacent words, rather than matching part of the phrase or or matching words separately.

Each word in the query is treated as a separate term (unless surrounded by double quotes) and by default a result only has to contain one of the terms.

This behaviour can be modified by using boolean operators (AND OR NOT) and round brackets e.g. to find mouse or human samples that do not have leukmemia you can use: NOT Leukemia AND ( mouse OR human ) .

Queries containing an asterix (*) or question mark (?) characters are treated separately. An asterix character will match any combination of zero or more characters e.g. leuk*mia will match to 'leukemia' and 'leukaemia' as well as 'leukqwertymia'. A question mark character will match any single characters e.g. m?n will match both man and men.

For certain attributes you can search for ranges of matching values. An example of this would be to find sample records that were last updated between the 1st and 4th of April 2014.

Attributes that can be searched using ranges are:

The search interface provides an easy way to filter the results based on attributes type and value. This is done using the facets available on the left side of the screen.

Each facet represents an attribute along with 10 of it’s most common values. Clicking on the facet title will filter the results of your query to sample records that contain the attribute with any value, whilst clicking on one of the facets in the sublist will filter the results more precisely to only show samples which contain the selected attribute type-value pair.

You can compose filters using multiple attributes or multiple values from the same attribute. Selecting different facets will produce an intersection of the results (an AND operator between the filters), whilst selecting different values within the same facet will generate an union of the results (an OR operation between the two filters).

NOTE: To apply the filters you need to click the Apply filters button at the top of the facet list

You can also create manual filters by following the reference guide on filters

Many samples use the project attribute do describe which project they are part of. Examples of this include the FAANG project, as well as HipSci and EBiSC. Therefore, to select samples from these projects we can filter on the project attribute. Note that if the facet for the project attribute is not available in the facet list (because only a maximum of the 10 most popular facets are displayed on the left hand side of the interface), you can hardcode the filter in the URL as explained in the reference guide on filters

To exclusively return samples within a specific project, select the project facet from the facet list (if available), or add filter=attr:project to the url after the search?. You can see the result of such filter using this link.

You should now be able to see the project facet on the left side of the page. You may select one or more values to further filter the results to a specific project[s]. Check out this link to see a search for FAANG and HipSci samples.

Many samples in BioSamples have external data associated to them which is stored in other archives such as ENA or ArrayExpress. If you want to limit your query to sample records that contain data in a specific archive, you can do it using an External Reference Data filter.

To see the most used external data archives, select the external reference facet from the list of facets available on the left side of the screen, or add filter=attr:external+reference to the url after the term search?.

Check out the result page for this query here

A list of available external archives will be available in the facet list. Click the archive you’re interested in, e.g. ArrayExpress, or add the corresponding value to the url, i.e. filter=attr:external+reference:ArrayExpress.

Check out the result page for this query here

You may have an ENA accession ID available but you need to retrieve the sample metadata associated to such accession from BioSamples.

The best way to do this is to use an external reference data filter. You can read the details about this in the filters reference page

Using facets is the easiest way of creating filters in the UI, but for more complicated searches it is usually easiest to manually add the filter to the url.

If for example you want to get all the samples associated to the ENA accession SRS359918, then just add filter=extd:ENA:SRS359918 after the search? part of the url.

Check out the result of this filter at this page

To check which are the most used archives in BioSamples, you can follow the procedure explained in the "Filter samples with data in an external archive" section.

You may need to get only the samples originated from NCBI, that have a accession pattern starting from SAMN. You can easily retrieve these samples using filter filter=acc:SAMN.*. You can either retrieve a list of accessions or all samples.