search

opensourceantibiotics / murligase

Everything to do with the Mur Ligase Project
29 stars 6 forks source link

Call for Open Source Antibiotics Fragment Contributions #13

Open danaklug opened 4 years ago

danaklug commented 4 years ago

Background: To address the need for new antibiotics, we have taken a fragment-based approach to discover hits against the Mur family of ligases, a target class which has been validated as essential to bacterial survival. In addition, these enzymes are absent from eukaryotes. A fragment library has been screened against Mur E and MurD as a starting point for optimization of small molecules against these targets.

What we have: Fragment hits from an initial screen against MurE and MurD, performed at Diamond screening facility, and a platform to screen additional fragment libraries or follow-up compounds. Further information about MurD hits can be found here. Further information about MurE hits can be found here. image

What we need: Additional chemical matter for screening. The Diamond screening platform is high-throughput and we would ideally be able to take full advantage of this. Some hits are currently being actively pursued (further information here, here, and here); others are not.

Since we’re at the early stage of the project, we need not worry too much about the contributed fragments being highly similar to the existing fragments. This is not a medchem hit-to-lead campaign, and the screened fragments can be quite different from what we have to date, but should ideally be of a higher molecular weight (i.e. 300-500 Da). Although we’d prefer fragments for screening, we are also accepting donation of interesting or unusual starting materials that can be easily coupled with building blocks to create new fragments.

Any donated compounds can be shipped to UCL, where fragments will be plated and sent to Diamond. Details about amount of compound needed and how to ship can be found here. Our first shipment will be at the end of November with future shipments every 1-2 months depending on compound availability.

Open source model: As this is an open source drug discovery platform, all data, including synthesis and characterization of donated compounds, will be shared in the public domain and there will be no protected intellectual property (i.e., no patents). An introduction to open source projects can be found here.

drc007 commented 4 years ago

Posted on LinkedIn https://www.linkedin.com/posts/chrisswain_call-for-open-source-antibiotics-fragment-activity-6600086794325282816-AIVE

and Twitter

https://twitter.com/OSantibiotics/status/1194320124863496192

pschmidtke commented 4 years ago

Hi Dana, can you give access to the (all) structures coming out of the Diamond screening?

danaklug commented 4 years ago

Hi Peter, All of the hits we currently have are detailed in the wiki pages linked above, and the initial dataset for all compounds tested in the original screen can be found in this spreadsheet. The plan for future screening sets is to openly share the structures of all compounds sent to Diamond and the screening results here on GitHub.

Hope that helps!

drc007 commented 4 years ago

@MFernflower Can you put them on the issue related to the particular cluster you had in mind.

drc007 commented 4 years ago

@danaklug @mattodd The article on LinkedIn has now been viewed over 1500 times and has 20 likes. Let's hope we get plenty of contributions.

bendndi commented 4 years ago

Hi all, @danaklug @mattodd It would be really useful if somewhere on the github page there was a spreadsheet of SMILES of all the compounds proposed to be made - kind of like a "molecule wishlist". is that something someone has time for and could curate as project moves forward? I'm not able to promise anything but I may be able to add Opensourceantibiotics compounds to the DNDi open synthesis network at some point in the futire (it's outside of our current disease scope). If I can do this that would add to the synthetic firepower. The aforementioned list would really help facilitate this. B

danaklug commented 4 years ago

@bendndi That sounds great! @fidiris and I can certainly put this together - maybe after we get the first round of data back would be best if that is okay timing for you. Thanks!

MFernflower commented 4 years ago

@danaklug @drc007 Just curious as to how the screening process works is it whole cell or a cell free assay?

Here's what I have proposed so far regarding this project: mfernflower_abx_frags

MFernflower commented 4 years ago

@danaklug @drc007 Do kindly note that compounds in green are marked as hypothetical because they might be neurotoxic - I cannot confirm or deny toxicity with out use of costly modeling packages or somebody able to interpret high-level QSAR - https://www.mdpi.com/1422-0067/20/9/2311/htm

(Update: one modeling paper says that removal of the cyclic nitrogen should abolish any neural actions but I do not want to take any chances asking for synthesis and antibiotic activity evaluation without confirming with several people that it would be safe!!!!!!!!!!!)

In general, a few of our compounds look like they might have neurological side effects but once again I might be pushing the cart before the horse!!!

danaklug commented 4 years ago

Thanks @MFernflower - a couple of comments:

Hope that helps!

MFernflower commented 4 years ago

Thanks @danaklug

I geuss this is a little out of my league as it's a totally different process than opensourcemalaria or mycetoma (optimization of known hitter)

Regarding concerns about neurotoxin - the compounds in green (especially so in hindsight) and a few compounds that have been screened/are in synthesis have some similarity to fentanyl so I have slight concerns for people handing these compounds in the lab https://www.mdpi.com/1422-0067/20/9/2311/htm

A paper says that with the green compounds my removal of the cyclic nitrogen should neuter neurotoxin but I am still concerned about it so I would not recommend synthesis untill we can weed those green compounds out further

bendndi commented 4 years ago

@MFernflower @danaklug I was just going to comment that the compounds in green look like they are designed to be fentanyl analogues. I was wondering what hypothesis you're using to propose them since I don't see how they are evolutions of any of the fragments identified . I'm probably missing something, can you just outline your throught process to come up with these as suggestions so I can understand :-)

I agree with @danaklug we are way way too early to start worrying or thinking about other potential activities of structures at this stage, we don't need to consider that until way later in the process and can start to engineer it out. The only way it could become an issue at this stage is if the synthesis (final compouds or intermediates) risks going near the structures of controlled substances (which these ones in green do) - and then the risk is not so much one of safety as it is one of legality.

bendndi commented 4 years ago

@MFernflower basically the concept we're looking at with these fragments is to grow them / merge them to move into more potent space. I totally recommend the review linked by @danaklug above, it is a great place to learn about FBDD since its written by a collective of the leaders in the field.

MFernflower commented 4 years ago

@bendndi @danaklug @drc007 I had just noticed that some of the murD ligase fragments look a bit like what I'd think would be an opioid receptor agonist (mostly the aryl-alkyl-amide portions of the molecules) I just elaborated and thought of a molecule that was some where in the middle of this (but that would not hit mu-receptor) - once again I normally do optimization - I am sorry for any inconvenience this caused and will gladly retract everything and stay within the mycetoma and malaria projects where I feel more comfortable

It's very hard being a autodidact as I simply do not have the intuition and experience of a professional pharma professional (I am trying however to get into school for biology)

I will gladly however read that FBDD paper!

danaklug commented 4 years ago

@MFernflower No worries! Just trying to give you a clearer idea of what we're trying to do with this project in particular. Please feel free to continue to contribute :)

MFernflower commented 4 years ago

@bendndi @danaklug I will still attempt to dock those to murD but again this is just because I noticed some similarities between a few fragments and opioid agonists (alas the latter being a target we do not want to hit). Again sorry to waste your time!

On Thu, Jan 23, 2020, 10:13 AM Dana Klug notifications@github.com wrote:

@MFernflower https://github.com/MFernflower No worries! Just trying to give you a clearer idea of what we're trying to do with this project in particular. Please feel free to continue to contribute :)

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/opensourceantibiotics/murligase/issues/13?email_source=notifications&email_token=AAYEWDULDT43MLNBYLSLON3Q7GXYBA5CNFSM4JMGJG72YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEJXV7SA#issuecomment-577724360, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAYEWDQCVWZ4KU2JWFWX6WDQ7GXYBANCNFSM4JMGJG7Q .

kym834 commented 4 years ago

Hi all, Kymberley here.

I wanted to introduce myself as I'm a new addition. I've just started with @alintheopen as a postdoc and am working on the Breaking Good Project. We are hoping to get involved in this project and make some fragments with undergraduate students in their labs this year.

I'm currently looking through the information on here to see what has already been done and coming up with some new possible targets for us to synthesise. Though, I was wondering if anyone has some suggestions? Maybe some areas of chemical space you want to explore but have been put on the back burner.

As we are working with first year students there is some limitation on what reactions we can do but please fire away with any suggestions!