Naka-pin na Tweet
Time to retire this account. It will stay up but I won't post and probably won't check it, either.
I'll be hanging out at @rapodaca@fosstodon.org for now instead. The posts at depth-first.com will continue.
English
Rich Apodaca
5.7K posts
Rich Apodaca
@rapodaca
Science, software, and startups.
United States Sumali Ağustos 2007
0 Sinusundan873 Mga Tagasunod
@andrewwhite01 @SuperScienceGrl In terms of expense, working on the wrong target. Zero effect, AFAICT.
English
What are the hardest problems right now in drug discovery? Has deep learning had any effect?
English
IMO, goal should be to ↓ bad interpretations, not "bad data." Readers should expect authors to provide the data needed to do that. How much & what kind depends on situation.
Bigger problem is how hard it is to turn this into editorial policy.
Corin Wagen@CorinWagen
Something I've wanted to write for a while—my argument for why journals shouldn't require 13C NMR for publication. (This post was inspired by the many long hours I've spent collecting 13C NMR spectra.) Hear me out! corinwagen.github.io/public/blog/20…
English
What do you do when doing something is worse than doing nothing? experimentalhistory.substack.com/p/the-rise-and…
English
When it comes to organic synthesis, ChatGPT may be kind of book smart, but it's not very street smart.
English
Here it confuses Oc1ccccc1 with benzene in the text, then proceeds to include the oxygen atom in the molfile, modulo one carbon, but with a complete connection table for the underlying graph, modulo one carbon, but with correct placement for oxygen.
English
🤣
My tinkering so far reveals that ChatGPT readily recognizes SMILES, always confidently decodes them, doesn't appear to be doing simple string matching, and has a ~50:50 average for simple cases.
I'm still blown away that this is a thing.
English
@cthoyt No, but I've poked around at it. It's interesting, but there's also a lot of centralization potential there that isn't recognized now because users are scarce. I'm not sure many people realize this.
English
@Dereklowe @OrdoSeclorum 1. No RSS auto-discover link (could trip up some readers)
2. Footer links to a page which links to science.org/blogs/pipeline… .
3. My reader (RssBook) sorts feed in _ascending_ chron. order (other feeds are reverse).
4. Feed title ("AAAS: Keyword...") is confusing.
English
I admit I laughed.
Sigma-Aldrich@SigmaAldrich
You've heard of Elf on a Shelf, now get ready for:
English
Canonicalization is one of those topics that always seemed like black magic to me. Here's an attempt to demystify. 🦀 #rustlang #cheminformatics depth-first.com/articles/2022/…
English
Endgame: robotic journals vacuum up the best papers from preprint servers, running circles around human editorial boards and reviewers for quality control, fraud detection, and relevance filtering.
English
@clemensisert Exactly the kind of paper I was looking for - thanks!
English
@rapodaca One approach of accounting for different tautomers is using data augmentation to train a model to predict the same experimentally observed value for all tautomers (tautomer invariance): nature.com/articles/s4200…
→ for inference, it doesn't matter if you choose the "right" tautomer
English
How do machine learning studies account for tautomerism? Not predicting the most stable tautomer and the like, but rather signaling that a molecular comes in tautomeric forms, and where?
English
English
Raising the question... what's the difference?
Paul Graham@paulg
For me one of the biggest surprises about current generative AI research is that it yields artificial pseudo-intellectuals: programs that, given sufficient examples to copy, can do a plausible imitation of talking about something they understand.
English