🦠 Model Request: Prediction of clinically relevant drug‐induced liver injury from structure using machine learning

[nataliyah123]

Model Name

Drug-induced liver injury prediction

Model Description

Prediction of clinically relevant drug-induced-liver-injury (DILI), based solely on drug structure using binary classification methods. The results presented here are useful as a screening tool both in a clinical setting, in the assessment of DILI as well as in the early stages of drug development to rule out potentially hepatotoxic candidates.

Slug

dili-pred

Tag

Metabolism, Toxicity, Cytotoxicity

Publication

https://pubmed.ncbi.nlm.nih.gov/30325042/,

Source Code

https://github.com/cptbern/QSAR_DILI_2019

License

None

[GemmaTuron]

Hi @nataliyah123 ! I have been doing some reading on DILI. There is this benchmark dataset from TDCommons that uses FDA-approved drugs. Do you know if there is overlap between it and the data you are using in this model? I am also linking here an interesting paper that cites more datasets, I'm not sure if they are open but would be interesting to investigate: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7702310/#R6

[nataliyah123]

hi @GemmaTuron I have attached the notebooks and dataset. Thanks a lot for the datasets links and will let you know.

[GemmaTuron]

Hi @nataliyah123

The notebooks look good, but before deciding next steps can you please:

• Check which datasets were used in the publication and whether these are overlapping with the ones in TDCommons (please note that SMILES might be given in different formats, so make sure to standardise the datasets before comparing them or you wont find matches)
• Write a short summary of the performances you have obtained with the different methods and compare to the ones obtained in the original publication

[Femme-js]

Hi @nataliyah123 !

Can you share the attachment for the training set here?

[nataliyah123]

hi @Femme-js Data and notebooks are all attached in the top comment. The file smileandactivity2.csv which I made for Zairachem is the new one, I just attached

[AdedejiAdewole]

hello, how can I work on this please

[GemmaTuron]

Please follow the outreachy contribution guidelines outlined by the Ersilia Community: https://ersilia.gitbook.io/ersilia-book/contributors/internships/outreachy-summer-2023

[GemmaTuron]

@miquelduranfrigola Do we want to finish the work on this model ¿

[GemmaTuron]

@leilayesufu If you finish your other tasks or are stuck, can start having a look at this model!

[GemmaTuron]

/approve

[GemmaTuron]

I've removed the notebooks from the issue as it disrupts the workflow:

Notebooks

Autogluon AutoML Dataset smilesandactivity2.csv

[GemmaTuron]

/approve

[GemmaTuron]

/approve

[GemmaTuron]

/approve

[GemmaTuron]

there were many modifications on the issue template that cause multiple fails in the workflows

[GemmaTuron]

/approve

[GemmaTuron]

Hi @leilayesufu I have tried too many times to change the model request back to its original form. I'll close this issue and please open a new one with the information provided, it'll be much easier. Thanks!

[GemmaTuron]

I am reopening this issue to check the new pipeline created by @GrantBirki when the approve command fails

[GemmaTuron]

/approve

[github-actions[bot]]

Workflow Failure ❌

@ (or other maintainers) the /approve workflow has failed. View the logs here for more information:

🔗 Workflow logs

You may need to delete the following repo that was created via this workflow run since the run was not fully successful: ersilia-os/eos8xjx

[GemmaTuron]

And that works just fine @miquelduranfrigola and @GrantBirki thanks!