297 karma · joined September 25, 2020
I personally have a clue, and the entire field of organic chemistry has a clue, given enough time and money most reasonable structures can be synthesized (and QED+SAScore+etc and then human filter is often enough to weed out the problem compounds that will be unstable or hard to make). Actually even some of the state of the art synthesis prediction models are able to predict decent routes if the compounds are relatively simple [0]. The issue is that in silico activity/property prediction is often not reliable enough for the effort to design and execute a synthesis to be worth it, especially because as typically the molecules will get more dissimilar to known compounds with the given activity, the predictions will also often become less reliable. In the end, what would happen is that you just spend 3 months of your master student's time on a pharmacological dead end. Conversely, some of the "novel predictions" of ML pipelines includign de novo structure generation can be very close to known molecules, which makes the measured activity to be somewhat of a triviality.[1] For this reason, it makes sense to spend the budget on building block-based "make on demand" structures that will have 90% fulfillment, that will take 1-2 months from placed order to compound in hand and that will be significantly cheaper per compound, because you can iterate faster. Recent work around large scale docking has shown that this approach seems to work decently for well behaved systems.[2] On the other hand, some truly novel frameworks are not available via the building block approach, which can also be important for IP.
More fundamentally, of course you are correct, and I agree with you: having a lot of structures is in itself not that useful. Getting closer to physically more meaningful and fundamental processes and speeding them up to the extent possible can generate way more transparent reliable activity and novelty.
[0] https://www.sciencedirect.com/science/article/pii/S245192941... [1] http://www.drugdiscovery.net/2019/09/03/so-did-ai-just-disco... [2] https://www.nature.com/articles/s41586-021-04175-x.pdf
1. that compounds derive from some difficult to handle gas is only a problem for the process chemists. In this case because the TFA-based precursor is available this is not a problem. Though technically, the production of that commercially available TFA will probably have involved F2.
2. Though indeed there are quite a few toxic compounds that incorporate fluorine, such as 5-fluoro-uracil, fluoroacetate etc this is not quite a general rule, as there are also plenty of non toxic medicines and materials that incorporate fluorine. It seems decently compatible with life compared, certainly compared to some other elements or classes of compounds generally.
3. specifically concerning nucleoside/-tide/-base analogs I wouldn't quite pin their cytotoxicity on fluorine, as a class these will tend to be toxic because they are similar enough to nucleosides to get where they need to be but dissimilar enough to cause problems and make systems that deal with nucleobase derivatives malfunction. Instead of 5-fluoro you can also put 5-iodo, 5-TFM, sulfur, alkynyl, etc there and get antiviral like, cytotoxic compounds
In any case, when it comes to this specific pfizer drug, I would say the presence of some fluorines is not a production problem and certainly not a toxicity red flag. For example, Derek Lowe, whose amusing post on some of the more dangerous fluorine chemistry you cite, also has the following to say about fluorine and medicinal chemistry:
"We med-chemists just love our fluorines - as long as we don't have to use, like, fluorine itself to make them - because they armor-plate parts of our molecules against being metabolized and can change the binding profiles of the parent structures like nothing else can."[0]
[0]https://www.science.org/content/blog-post/new-ways-fluorinat...
It is possible to do asymmetric reductive amination using enzymes but this is out of the scope of clandestine producers. Likewise a chemical and not enzymatic asymmetric reductive amination would be easy in a bench lab, probably to expensive and impractical in a clandestine setting.
[0] Joseph S. Bozenko, "Clandestine Enantiomeric Enrichment of d–Methamphetamine via Tartaric Acid Resolution", JCLIC, 2008, vol.3 (not publicly accessible but you can find this if you know where to look)
the shake and bake method uses "sudafed"/pseudoephedrine. therefore, it will produce d meth
> This is definitely true for meth, where almost every method yields racemic meth
not quite true, because the main precursor pseudoephedrine already had the correct stereochemistry in place. you would actually have to do effort to racemize that asymmetric carbon. But it is true that if your precursors are racemates or not asymmetric and you are not using some fancy asymmetric catalysis or tedious resoltuion your product will be racemic. An example of a racemic meth synthesis that does not involve P2P is direct amination of allylbenzene.
[0] https://www.izotope.com/en/learn/mastering-for-vinyl-tips-fo...
edit: and re:KYC, there is a money limit above which you need to provide KYC/registration. This limit is 24000 CZK = 1096 USD
I'm not saying the conclusions of Bellingcat et al are necessarily wrong (though personally i take some of what they are writing with a grain of salt), but I would doubt that open source intelligence groups are necessarily that anti systemic or a challenge to state directed information flows and embargos. Finally, here is a fairly interesting article with a critical view of the Bellingcat group [2], though the source, Mintpress, is quite biased so critical reading is advised.
[0] https://www.bellingcat.com/about/
[1] https://www.bellingcat.com/news/uk-and-europe/2018/10/09/ful...