RSS Amplifier

HP Erskine · Jun 16, 2026

My preprint on orexin receptor ligands

0
Sign in to vote or save

HP Erskine · HP Erskine

Yay, I have a preprint!

Now that I know how to do some computational chemistry, my next target is global warming. Namely, an enzyme in cow guts that makes methane. I’m going to look for a good inhibitor using a similar pipeline as the one I used for this project.

This is confirmation that I can sit in my bathrobe with a nice fat support cat on my lap and do—hopefully—helpful chemistry on my computer from home. A dream come true. No lab materials were wasted in this study.

For this study, I was motivated to explore newer insomnia medications. I screened 80 compounds that are similar in structure to those that act on the brain’s orexin receptors, a system that governs sleep and wakefulness. I was pleased to get confirmation that some of my best candidates had already gained some interest in the literature.

I hope this means that my computational pipeline is a good one. Here’s a link to my pipeline.

If I could make any changes, I would use a virtual receptor docking system that is more flexible. However, the scope of my study was for a class project, which didn’t allow for that sort of broader exploration.

Why did I chose to study the orexin system?

I hate insomnia drugs. I worry that many increase the risk of dementia. (There’s a terrifying wakeup-call study that shows that ambien (zolpidem) shuts off the glymphatic system in mice at least—that’s the brain’s waste clearance system.) I want to come up with safer alternatives. My mom, like many people with dementia, suffered terrible insomnia and for a while took drugs that worked through similar (GABAergic) mechanisms. I am beyond motivated to tackle this problem.

The orexin system may not be the answer to this problem.

But this is, at least, information that might help find a better answer, somewhere.

Here are some figures from my paper; the first gives you an idea of the steps involved.

The steps in my computational pipeline. Everything was done in Python except the last step, where I used R to do what R does best: statistical analysis.
A. My four seed compounds. Three are drugs on the market, known as DORAs, which I gather no one uses because they are too expensive and insurance does not cover them. The fourth, oveporexton, is my only agonist, and currently (2026) under review as a treatment for narcolepsy. B. Two compounds my pipeline uncovered as promising orexin receptor binding candidates.
I am in love with Principal Component Analysis (PCA), such a powerful statistical method to find patterns in great heaps of otherwise unmanageable data!
More data…

Read the original on hperskine.substack.com

Comments

Nothing yet. Say the first thing.

    Sign in to join the conversation.