API Examples

Graphics

Gain/loss frequency chromosomal ideogram

http://progenetix.org/api?db=progenetix&collection=samples&api_out=ideogram&icdm_m=8144/3,8140/0

• … produces a chromosomal ideogram (api_out=ideogram) with gain/loss frequencies, for cancer samples with ICD-O codes 8144/3 and 8140/0 (icdm_m=8144/3,8140/0) selected from the progenetix database (db=progenetix)

Gain/loss frequency histogram

http://arraymap.org/api/?db=arraymap&collection=samples&api_out=histogram&icdt_m=c18,c20&randno=100

• This query will produce a histogram of regional gain/loss frequencies in samples matching the loci icdt_m=c18,c20 (e.g. c189 - large intestine …, c181 - appendix & c187 - sigmoid …) and “c20” (rectum) from arrayMap (db=arraymap). Also, the query will return only 100 random samples (randno=100) out of all matches.

Example histogram of cancer genome imbalances in 100 random colorectal carcinomas, directly generated from the API call shown

http://arraymap.org/api/?db=arraymap&collection=samples&api_out=histogram&icdm_m=0000&anysegmaxkb=10000&randno=1000&maxy=25&laby_m=0,7.5,15,22.5

• This query will produce a histogram for the frequency of germline CNVs, based on 1000 random reference arrays from the arrayMap collection. As a simple quality filter, only arrays are considered which lack any single segment >10Mb (which would be pretty much a marker for somatic changes or pathologic germline variants, though we found once a >10MB segment w/o apparent phenotype ourselves.

Geographic map of publications

http://www.progenetix.org/api/?imgw=1024&api_out=pubmap&locscale=1&api_scope=publications&mapcaption=1

• A geographic map of all publications registered in the Progenetix publication database, with locations derived (mostly) from the corresponding authors’ institutions, and marker sizes corresponding to the number of samples reported from there.

Data/Text

Called CNA segments from selected samples =api_out=segments=

http://progenetix.org/api/?db=progenetix&collection=samples&api_out=segments&text_m=metast&icdm_m=8500/3

• The api_out=segments returns a list of CNA segments from infiltrating duct carcinoma samples (icdm_m=8500/3) with a “metast” word stem match (text_m=metast), e.g. having “metastasis” or “metastasised” somewhere in their annotations. As the list shows, these are only very few out of the expected number of samples …

Tab-delimited data table from selected samples =api_out=samples=

http://arraymap.org/api/?db=arraymap&api_out=samples&icdm_m=81&exportk_m=UID,DIAGNOSISTEXT,ICDMORPHOLOGY,ICDMORPHOLOGYCODE,ICDTOPOGRAPHY,ICDTOPOGRAPHYCODE,GRADE,TNM,DEATH,FOLLOWUP&randno=200

• … results in a tab-delimited table with columns representing the comma-concatenated exportk_m parameters, from 200 (randno=200) adenocarcinomas and -adenomas (icdm_m=81) of various histologies

JSON data file from selected samples =api_out=samples&api_doctype=json=

http://progenetix.org/api/?db=progenetix&api_out=samples&api_doctype=json&icdm_m=817&randno=20

• … results in a JSON data file, a wrapper containing the calling information and the =data[]= attribute with sample objects inside

Sample count =api_out=count&api_doctype=json=

http://arraymap.org/api/?db=arraymap&api_out=count&genome=hg19&api_doctype=json&icdm_m=8,9&locus_m=17:7512444-7531593:-1

• … results in json data structure (api_doctype=json) containing the call information as well as the count of all neoplasias (icdm_m=8,9) with a deletion overlapping the TP53 CDR (locus_m=17:7512444-7531593:-1). Avoiding the json specification will just return the number of hits:

http://arraymap.org/api/?db=arraymap&api_out=count&genome=hg19&icdm_m=8,9&locus_m=17:7512444-7531593:-1

ICD morphology list =api_out=icdmlist&text_m=breast=

This example uses a text match to list all ICD morphologies having samples with a match in any annotation field.

http://arraymap.org/api/?project=progenetix&genome=hg18&db=progenetix&text_m=breast&api_out=icdmlist

Looking up all ICD morphologies with samples in colon or rectum (C189 and C20):

http://progenetix.org/api/?api_out=icdmlist&icdt_m=189,20&db=arraymap

Looking up all publications for sample numbers and geographic data:

http://progenetix.org/qmongo/?db=progenetix&collection=publications&all=multvalues&afqfield=PMID&querytext=...&exportk_m=PMID,NO_ACGH,NO_WES,NO_WGS,CITY,COUNTRY&sortby=COUNTRY&counter=y

---

Michael, 2018-07-20