Pat Sharp's Avatar

Pat Sharp

@patsharp.bsky.social

Drug developer | Co-founder of Gate Bio | Fmr Big Pharma | Inventing small molecule medicines | Artist | Polymath

536 Followers  |  1,255 Following  |  149 Posts  |  Joined: 03.12.2024  |  2.016

Latest posts by patsharp.bsky.social on Bluesky

Post image

Let’s Go!!

25.07.2025 00:37 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0
Post image

www.genengnews.com/topics/drug-...

#chemsky #biotech

08.07.2025 23:00 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0
Post image

Home

07.07.2025 04:14 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Post image

Chat can use the RDkit #chemsky

07.07.2025 04:12 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

Everyone wants to meet on Zoom again, rather than in person.

Has the promise of the return to office in-person productivity and relationship building failed to deliver?

19.06.2025 05:22 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

I got this same ad

07.06.2025 22:07 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

Lots of really good references go a long way. You’re not usually hiring most people to give presentations. If the science is good but the presentation is bad (nervous speaking, poor slide formatting etc.) but everyone says they worked with says they are a rockstar then they’re probably amazing!

05.06.2025 10:34 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

The origin from chirality - from amino acids to pasta 🍝

31.05.2025 11:49 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Preview
Harvard’s former chemistry chair takes new position at Chinese university Charles Lieber, the nanoscience pioneer convicted for not disclosing his ties to China, has just joined Tsinghua University

www.chemistryworld.com/news/harvard...

#chemsky

03.05.2025 10:54 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

Just got around to trying. Was excellent for single compounds but failed on batches of 2 or more.

28.04.2025 04:49 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Post image Post image Post image Post image

The Shine Dome - home of the Australian Academy of Science - Canberra ACT.

It’s my favourite building in Australia.

I’d love to give a seminar here one day.

#chemsky #australia #science #architecture

28.04.2025 04:00 β€” πŸ‘ 4    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

My favorite topic

19.04.2025 05:11 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Post image

Presenting on Molecular Gates at the RACI MCCB conference 2025.

#chemsky #chembio #biotech #pharma

19.04.2025 04:22 β€” πŸ‘ 3    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

#chemsky

16.04.2025 21:32 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Post image

Chat getting better at organic molecules.

Asked for alprazolam. Got the chlorine atom. Almost got the benzodiazepine and almost got the triazole.

How many months until it starts to nail it?

16.04.2025 21:32 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

#chemsky

16.04.2025 16:08 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

In real molecules, most atoms follow the octet rule and thus tend to have even degrees (e.g. carbon has 4 bonds, oxygen often 2, etc.), so a lot of molecules will be Eulerian (0 odd-degree nodes) β€” and therefore drawable in one continuous stroke.

16.04.2025 16:08 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

This insight might help automate or optimize molecular rendering software β€” like for generating SMILES strings, kekulΓ© structures, or 2D depictions that are human-readable and efficient.

16.04.2025 16:08 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

Euler trail rule:
β€’ 0 odd-degree nodes β†’ 1 loop
β€’ 2 odd-degree nodes β†’ 1 line
β€’ n odd-degree nodes β†’ n/2 lines

Every drawing stroke has to start and end somewhere β€” the math tells you how many strokes you’ll need.

16.04.2025 16:08 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

Graph theory shows up in the weirdest places. In this case, the path to a 2D chemical diagram runs through Euler.

If you’re designing cheminformatics tools, molecule editors, or even just teaching, this rule is a gem.

16.04.2025 16:08 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

And yeah β€” I later found out Euler figured this out in 1736 to solve the Seven Bridges of KΓΆnigsberg problem.

But it’s still wild to rediscover it from scratch β€” while thinking about molecule drawing.

16.04.2025 16:08 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

Eventually I realized: the number of non-overlapping lines you need to draw a molecule is just:

(# of odd-degree atoms) / 2

That’s it. Independent of the even-degree atoms. Totally general.

16.04.2025 16:08 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

I noticed that every time an atom has an odd number of bonds (say 1 or 3), you can’t draw through it cleanly β€” it must be the start or end of a line.

So a molecule with two of these odd-bonded atoms needs one line to connect them.

Four odd atoms? You need two lines.

16.04.2025 16:08 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

Here’s the idea: each atom in a molecule is a node, and each bond is a line connecting them β€” i.e., an edge.

This makes any chemical structure a graph. Which means we can borrow some tools from graph theory.

16.04.2025 16:08 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

I was thinking about how we draw organic molecules β€” those 2D skeletal structures we all know β€” and I stumbled into an old mathematical idea that made me stop in my tracks.

Turns out, the number of continuous lines you need to draw any molecule can be predicted.

16.04.2025 16:08 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

#chemsky

07.04.2025 05:44 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Post image

50g ATP processed in a few SECONDS…gone…let that sink in.

Muscles

07.04.2025 05:44 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

Glad I’m living back in Aus again

amp.abc.net.au/article/1050...

04.04.2025 09:50 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

It was an incredible place to learn medchem

01.04.2025 23:23 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Preview
From Cortisone to Enlicitide: A Journey of Synthetic Chemistry Innovations at Merck Necessity is the mother of invention. Most synthetic chemistry innovations are driven by our desire to make molecules. In the first half of the 20th century, much of this work was inspired by natural ...

One thing you can count on from me is to share great synthesis. While we may be competitors, I’ll highlight this review of #MerckChemistry innovations in #JOCasap from my respected industry colleague and friend @drlcsquare.bsky.social The work here is inspiring and drives us to our best #ChemSky

01.04.2025 22:42 β€” πŸ‘ 37    πŸ” 10    πŸ’¬ 4    πŸ“Œ 0

@patsharp is following 20 prominent accounts