I launched myself into my graduate studies from a deep fascination of plant secondary metabolites. After eight years of chemical genetics my interest in bioactive small molecules from plants was deepened.
It was, therefore, a logical choice to work on my favorite plant and contribute my innovation, energy and scientific knowledge to the medicinal cannabis industry.
Sonoma Labworks Summer 2015
I was fortunate to have my start in the cannabis industry at Sonoma Labworks in 2015. Although brief it was an influencial exposure to the inner workings of a succesful cannabis analytical laboratory as well as large scale greenhouse cultivation for the California cannabis brand Absolute Xtracts. We spent a signifcant amount of our time performing quality control of vape products and Care by Design high CBD products, which were very rare in the market at the time.
Back then, and still now, I was concerned with reproducible research tied to visual representations of data for insight and consistency using the programming language R.
To that end I was able, in a short time, to whip up a few R scripts that was able to graph raw data from the High Pressure Liquid Chromatography (HPLC) and Gas Chromatograph (GC) instruments. The data from these instruments are the bread and butter of any cannabinoid and terpene potency testing lab.
A problem with instruments such as these is that they have to be calibrated often, and cannabinoids such as THCA degrade at room temperature over time. Therefore, it is not as trivial as leaving a mixed cannabinoid calibration vial in the instrument to inject at a regularly. Another issue that arises with HPLC is day-to-day drift in the chromatogram elution windows and inter day variation. This can be problematic when attempting to detect multiple analytes that are very similar and closely eluting. Nonetheless, a majority of these issues have been ameliorated with the more expensive ultra-high pressure chromatography (UHPLC) platform.
What the open source software R can do for the scientist concerned with reproducible research is make it easy to visualize large amounts of data and keep detailed records to detect and follow problems such as daily or day-to-day variations.
Square one is to look at how samples during the working day and long operation runs of the HPLC deviated from the calibration standard, shown here below showing tight windows for peak detection for the cannabinoids of interest (not labelled).
This principle is applicable as well to more busy data such as the chromatogram from the GC. In gas chromatography (GC) many terpenes elute or travel closely together because they are similar molecules. Anyways, you can see a re-plot of a GC calibration curve below and judge for yourself.
My overall vision was to have a cannabinoid and/or terpene chromatograms printed directly onto the certificate of anlysis (COA) along with the most recent calibration curves, but my aspirations and deep need for formalism took me elsewhere (keep reading for the details).
Bloomfield Industries INC pre-licensing application 2015
Back then the cannabis market in California and much of the West coast was operating under a patchwork of Caregiver licensing. This was unsettling to me as regulations were unclear with no clear path to Federal compliance at the time.
Seeking new opportunities to work with State mandated medical Cannabis I reached out to my network and got a lead...
Big changes were happening in the East coast with the launch of the New York Medical Marijuana program in 2016. The program was a state sanctioned initiative championed by Gov Cuomo and tightly regulated by the Department of Health (DOH). My interest was piqued and I was contracted to write the Research and Development section of the pre-licensing application submitted to the DOH by Bloomfield Industries Inc a Staten Island, NY startup.
I can share with you my contribution below since, for whatever reason, it was the only portion not redacted from the final approved application. Know that I consider the write up quite rudimentary, but keep in mind my audience was likely a layperson reviewer with a score card and a pencil at the NY DOH. Therefore, I thought my gusto to be more "academic" may have not impressed the reader.
Bloomfield Industries INC Fall 2015-Spring 2016 (Staten Island, New York)
More details comings soon...