UNC Chemical Dictionary
A purpose-built resource for drug checking. Always public, always free. Brought to you by the UNC Street Drug Analysis Lab.
🚧 Work in progress. We are building this out and sharing a preview with y'all now. If you see anything you think could be said or classified better, please let us know. Some of this information will change as we refine it. We are sure there are mistakes, and there are dozens of substances that haven't been tagged yet. So, if something is awry, trust your instinct! We are sharing this now because of so much community demand. And, also, we want your feedback.
Individual chemical names can be hard to understand, and there are many classes of compounds found in street drugs. Further, existing drug classification systems were not designed to communicate health harms. Therefore, we are building an in-house chemical dictionary that classifies substances so that they can be understandable for public health. Additionally, in data analysis, we have only rudimentary capabilities to identify classes of molecules. Building the ChemDict will allow us to gain deeper insights into our data.
What we built
The chemdict allows us to classify substances we identify into pharmacological classes. This allows for: (1) data analysts to use classes of molecules instead of solely relying on compound names; (2) helps standardize chemical naming conventions; (3) allows us to link our results with other chemical databases using PubChemCID, SMILES, etc. The chemdict will contains basic plain English descriptions of substances than can be used to provide simplified information back to sample donors, and can be incorporated into apps, reporting tools, dashboards, etc.
Who we built for
🤓 Data Analysts Epidemiologists or data scientists analyzing machine-readable data from drug checking and want to group substances by chemical class. Could be applied to UNC drug checking data, but also data from other drug checking services or other sources.
💗 Harm Reduction Programs Program staff who need to communicate results concisely to sample donors will find 6-word descriptions handy. Additional information contained in notes is focused on biomedical info right now, but could be expanded if we crowdsource good harm reduction resources.
🧪 Startup Lab Services Other chemistry labs who are starting to do lab-based drug checking and can use synonyms to align data output.
💻 Software Developers Who are building data viz off of our data (or their own) and can use our categorization scheme and sixwords to make sense of the drugs that they encounter. (example)
Who built this and how
Anuja Panthari, Nabarun Dasgupta, Meredith Neiss, Nate Erskine, Erin Tracy (add Jalice Manso, others...)
Resources
Where these sample come from
Run to view results
Run to view results
Tag frequency
This is the number of substances for each tag. It is not the number of samples.
Run to view results
Run to view results
ChemDict
The ChemDict comes in two version: Concise and Exploded. This refers to how the tags are recorded, either as a delimited (;) list or as more than a hundred 1/0 columns (one for each tag). Concise is best for those who are concerned about data efficiency, while Exploded will suit the needs of data analysts who want to get going quickly.
Run to view results
Concise Version
Run to view results
Run to view results
Exploded Version
The tags are separated by semi-colon (;) that should be easy to parse. To save you a step, in this "exploded" version of the ChemDict, each tag is its own column taking values of 1/0/. for yes, no, and missing.
Run to view results
Here's what the tags in the Exploded ChemDict look like with 5 random substances. The other identifiers (PubChem CID, etc.) are included, but does not currently contain vernacular, synonyms, and notes due to formatting issues we are working on. But still, this is pretty useable when importing into SAS, Stata, Python, etc. 🤷🏾 Hey, this shit is hard to organize, so we do what we can!
Run to view results
License
You can download and use the ChemDict however you'd like. For the sake of accountability, we suggest noting the date on which you downloaded the ChemDict as part of your citation.
Copyright (c) 2023 UNC Street Drug Analysis Lab
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.