Integrating coding into chemistry helps science progress faster

 High-throughput chemists like my team and I usually notice the inefficiencies the most. Average lab chemist in a day he takes 8 he LC-MS samples evenly spaced he quickly gets used to the imperfections of the 20 second workflow and usually doesn't complain . But when a high-throughput chemist processes 96 samples at a time from a microplate, the small limitations can add up to become a major annoyance or even a hindrance. At some point in my career, I was asked to sort his LC-MS data using a specific workflow. This process involved remoting to multiple machines and performing tedious workarounds to combine file and batch export data at high throughput. Exporting analytical data was slow, so if we later noticed new by-products, we had to weigh the usefulness of identifying by-product yield patterns across the disk against the value of the time to redo the export process. Information was lost regularly as the benefits were often not immediate. When designing software, it's obvious that you don't do that. But we were chemists, not software designers.

But a year before he did, I attended a free introductory programming course. The company was run by an acquaintance of mine from college, and my first job in the industry was as a chemist, so I thought it would be a great idea to work alongside my job. I didn't think he had much reason to code, but he seemed like he could be useful one day.

i was right My immediate memory was limited to all Python insight when I was shown a complex LC-MS workflow. But the most important thing I learned from this course was that it is possible. I knew Python could fix the analysis export, so I googled almost every line until I figured out what I wanted to do and created a working process. It wasn't well written and didn't take advantage of many useful open source packages that I didn't know about, but it was enough. The new script meant my team and I were free to collect any information we identified.

#chemistry #science #experiment