Application Programming Interface (API) For MICHA
MICHA provides API to give programmatic access over FAIRified drug screening protocols, target data profiles and disease indications for 0.8M compounds.
So far, our annotation team has FAIRified 11 protocols including 5 cancer protocols and 6 covid19 protocols. But we will FAIRify more protocols in future. Altogether, MICHA FAIRified 5,605 compounds and 1,377 cell lines. This information can be programmatically extracted and integrated with other public tools using API under FAIR principals. Users can extract data using various options and apply different filters to get specific results. Following are some of the examples on how to extract FAIRified protocols.
Users can extract information from any of the FAIRified protocols using compound names. Compound names are case insensitive. Following link returns FAIRified drug screening data for gefitinib, which is an approved EGFR inhibitor.
For querying multiple compounds users need to enter compound names separated by comma. Following example returns information for three compounds (gefitinib, afatinib and apatinib)
Though approved compounds have proper names but there exist millions of preclinical or investigational compounds that are not yet assigned with proper names. These compounds are usually referenced using standard identifiers such as smiles or InChiKeys. Therefore, MICHA also support querying FAIRified compounds using standard InChiKeys. Following link returns FAIRified drug screening data for gefitinib using its standard InChiKey.
Users can also query FAIRified cell lines using cell line name as shown in the following link:
Furthermore, users can also apply multiple filters separated by ‘&’ operator. For instance, FAIRified compound across a particular cell line can be obtained as.
Other than compound or cell line information users can also query FAIRified annotations using thresholds on minimum or maximum concentrations. For example, following link can return all FAIRified compounds and cell lines with concentration: 1-1000 nM. We adapted nM as standard activity units in MICHA.
There are several activity metrics used in different protocols such as IC50, EC50, DS and AC50. Users can also filter data using various metrics. For example, following link returns ‘AC50’ data between concentration: 1-1000 nM.
Following are the links to get FAIRified drug screening annotations for different detection technologies.
Other than aforementioned search queries, users can also query FAIRified drug screening data using various other options such as cell density, plate types, volume per well, vehicle etc. API links for these searches are shown below:
There are potent target profiles for ~0.8M compounds in MICHA. These target profiles can be extracted or linked with external tools
using FAIR principals. Furthermore, same API link also returns tested disease indications for the compounds as well
as physiochemical properties. This information can be fetched using standard InChiKeys.
Following link returns primary and other potent targets for gefitinib as well as disease indications that are currently under trials at: https://clinicaltrials.gov/. https://api.micha-protocol.org/get_targets/XGALLCVXEZPNRQ-UHFFFAOYSA-N
Users can also query more than 1 compound in one request. Micha allows target profiles for maximum of 500 compounds in one API request.
However, users can fetch thousands of compounds using loops in the code. For querying multiple compounds users need to enter standard
InChiKeys separated by comma. Following example returns compound target profiles for three compounds (gefitinib, afatinib and apatinib)