Final Preparations for Paper Submission

[mattodd]

Hi all. I hope everyone's well. An update on the paper (this issue closes #26).

The current draft is here, along with all the other files. We've incorporated some email comments from people who are not regularly on Github, which has been useful.

We're just waiting on a couple of things that lockdown interrupted:

1. Validation of the potency of the final four active compounds. They've been measured in duplicate in Dundee, but Dundee's (perfectly reasonable) policy is that published potencies should be measured in triplicate. Completing this should be quick after lockdown.

2. Confirmation that the more recent compounds in the paper share the same phenotype, i.e. that they are all still active in the ion regulation (PfATP4) assay. I'm 95% sure we've not scaffold hopped, but it's a loose end. This means shipping compounds to Adele and Kiaran in Canberra, and the temporary snafu is that the compounds are currently locked away in our shutdown UCL lab. We'll get them out as soon as we can.

For both these final measurements we have stocks of the compounds, which is a positive, and both relevant assays are up and running.

In the meantime we can get the paper ready to submit. This current draft is looking very nice (don't worry about the reference list - @edwintse is going to format that last thing). I need people to have a read and suggest any changes they'd like. Ultimately everyone will need to give the green light to submission, and that can be given now if you want. We're going to target Nature Comms since there's a lot of unusual work here.

The SI is also here, and it would help if people could also quickly verify their contribution to this document.

I also need people to check (at end of paper):

1. Competing interests (we currently have none)
2. Manuscript contributions (we've made a decent guess)
3. Any additional funding that needs to be declared (specific to the project, rather than e.g. core company funding)
4. Any co-authors we may have missed (this is super important, and it's easy to make a minor mistake when there are lots of authors)

Suggestions for referees for the paper would be very useful too.

@jonjoncardoso @spadavec @wvanhoorn @btatsis @BenedictIrwin @gcincilla @holeung @alintheopen @IamDavyG @sladem-tox @luiraym @murrayfold

[mattodd]

Potential new paper to cite if we want to highlight the importance of human-AI collaboration.
https://pubs.acs.org/doi/10.1021/acs.jmedchem.0c00163

[BenedictIrwin]

The manuscript looks great, thank you for preparing this. Under competing interests I would normally write "BWJI is an employee of Optibrium Ltd." just to make it absolutely clear, but it's not necessarily a competing interest in this case. I will check with other authors linked to our entry who may not see this post.

[adw62]

This looks good, awesome work. I have no competing interests and would add my funding as 'ADW would like to acknowledge the EPSRC Centre for Doctoral Training in Computational Methods for Materials Science for funding under grant number EP/L015552/1'

[gcincilla]

Well done guys! Thanks for pushing this forward. The manuscripts it's very nice and it reads fluently. Targeting Nature Comms is certainly a challenge, but if they can appreciate the innovation represented by the OSM modus operandi, it can be accepted smoothly.

There are no competing interests or additional funding that needs to be declared from my side.

A part from this I realized that in my models description in tables 1 and 2 I don’t mention which descriptors I used, as other entrants did. So, if it’s still possible, I would like to include such information, modifying the my models description in the following way:

• Table 1: "PfATP4 Ion Regulation Activity classification model using: CDK descriptors, ECFC4 fingerprints and Random Forest."
• Table 2: "Pfal inhibition classification model using: CDK descriptors, ECFC4 fingerprints and logistic regression (with: stochastic average gradient as solver, uniform regularization and learning step size = 0.01)."

@mattodd, is this still possible?

Reference for CDK (if applicable) is: Steinbeck et al. The Chemistry Development Kit (CDK): An Open-Source Java Library for Chemo- and Bioinformatics. J. Chem. Inf. Comput. Sci. 2003 Mar-Apr; 43(2):493-500, doi:10.1021/ci025584y

Finally, is the Supplementary Information section accessible?

[edwintse]

@gcincilla Of course! We'll make the additions. Are you not able to see the SI file from the link above?

[gcincilla]

Perfect @edwintse. Thank you also for pointing me at the SI file, I misread the original message.

[murraynrobertson]

It reads very well. Great job pulling all this together

Funding addition for me

"MNR would like to acknowledge the EPSRC Future Continuous Manufacturing and Advanced Crystallisation (CMAC) Research Hub (Grant No. EP/L015552/1)"

[spadavec]

Hello @edwintse

We have no funding notes to add. However we noticed that our entire model description is in the table of contents of the SI--is that by design?

[edwintse]

@spadavec Oh, that's not intentional so thanks for pointing that out! Your model description had the Heading style so it was picked up by the ToC generator. I'll fix it up.

[BenedictIrwin]

I have collected the following suggestions via email:

Under competing interests (if we are including all company affiliations):
MÖ is an employee of Optibrium Ltd.
BWJI is an employee of Optibrium Ltd.
TMW is an employee of Intellegens Ltd.
GJC is a Director and employee of Intellegens Ltd.

Under funding:
GJC acknowledges financial support from the Royal Society

After a bit of email discussion with the Intellegens/Optibrium group there are a few more small edits to the SI section if possible:

1. The SI title "Optibrium’s Model" -> "Optibrium/Intellegen’s Model" or similar for consistency with the table in the main text.

---

2. Inclusion of the following recent citations in the SI section, with the opening line being changed to:

"Machine learning was performed by the Alchemite tool (developed by Intellegens and available through Optibrium) [1,2,3] ..."

[1] Practical Applications of Deep Learning to Impute Heterogeneous Drug Discovery Data
B.W.J. Irwin, J. Levell, T.M. Whitehead, M.D. Segall & G.J. Conduit
J. Chem. Inf. Model., 2020, 60, 6, 2848–2857

[2] Imputation versus prediction: applications in machine learning for drug discovery
B.W.J. Irwin, S. Mahmoud, T.M. Whitehead, G.J. Conduit & M.D. Segall
Future Drug Discovery 2 (2020)

[3] Imputation of Assay Bioactivity Data using Deep Learning
T.M. Whitehead, B.W.J. Irwin, P.A. Hunt, M.D. Segall & G.J. Conduit
J. Chem. Inf. Model., 59, 1197 (2019)

They were also keen to include the reference(s) in the main text table if possible otherwise it won't be clear what "deep imputation" refers to, as it's a bit of a custom/novel method.

---

3. End of second paragraph:

"compounds tested in are easily modelled"

change to

"compounds tested in those labs are more easily modelled"

---

4. Start of third paragraph:

"Alchemite produced accurate error bars for both training and test results."

changes to

"Alchemite specializes in estimating error bars and produced these for both training and test results."

Sorry for these changes at a late stage. Thanks!

[wvanhoorn]

Is there mileage in adding the OrcID of authors who have one (yes I do have one, never used since I don't publish much)

…

On Wed, 24 Jun 2020 at 12:14, Benedict Irwin ***@***.***> wrote: I have collected the following suggestions via email: Under competing interests (if we are including all company affiliations): MÖ is an employee of Optibrium Ltd. BWJI is an employee of Optibrium Ltd. TMW is an employee of Intellegens Ltd. GJC is a Director and employee of Intellegens Ltd. Under funding: GJC acknowledges financial support from the Royal Society After a bit of email discussion with the Intellegens/Optibrium group there are a few more small edits to the SI section if possible: 1. The SI title "Optibrium’s Model" -> "Optibrium/Intellegen’s Model" or similar for consistency with the table in the main text. ------------------------------ 1. Inclusion of the following recent citations in the SI section, with the opening line being changed to: "Machine learning was performed by the Alchemite tool (developed by Intellegens and available through Optibrium) [1,2,3] ..." [1] Practical Applications of Deep Learning to Impute Heterogeneous Drug Discovery Data B.W.J. Irwin, J. Levell, T.M. Whitehead, M.D. Segall & G.J. Conduit J. Chem. Inf. Model., 2020, 60, 6, 2848–2857 [2] Imputation versus prediction: applications in machine learning for drug discovery B.W.J. Irwin, S. Mahmoud, T.M. Whitehead, G.J. Conduit & M.D. Segall Future Drug Discovery 2 (2020) [3] Imputation of Assay Bioactivity Data using Deep Learning T.M. Whitehead, B.W.J. Irwin, P.A. Hunt, M.D. Segall & G.J. Conduit J. Chem. Inf. Model., 59, 1197 (2019) They were also keen to include the reference(s) in the main text table if possible otherwise it won't be clear what "deep imputation" refers to, as it's a bit of a custom/novel method. ------------------------------ 1. End of second paragraph: "compounds tested in are easily modelled" change to "compounds tested in those labs are more easily modelled" ------------------------------ 1. Start of third paragraph: "Alchemite produced accurate error bars for both training and test results." changes to "Alchemite specializes in estimating error bars and produced these for both training and test results." Sorry for these changes at a late stage. Thanks! — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#28 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADCV5KSJSOUG7KLSTJSBJVDRYHNYZANCNFSM4OBRXY4A> .

-- *Willem P van Hoorn, PhD* Chief Decision Scientist *Exscientia* | http://drug.design The Schrödinger Building Oxford Science Park, Oxford, OX4 4GE, UK email: wvanhoorn@exscientia.co.uk phone: +44 1865 818941 Please note that this email communication and any attachments to it may contain confidential or privileged information and so circulation should be restricted to the addressee’s organisation. If you are not the addressee, please let us know by return email, then delete all copies of this email. Personal data that you share with us will be stored and used in accordance with our Privacy Policy which can be accessed on our website. Exscientia Ltd is a company incorporated in Scotland. Company Reg. No SC0428761. Registered Office: Dundee One, River Court, 5 West Victoria Dock Road, Dundee, DD1 3JT