October 8, 2020. 6 - 7 PM CEST
During this webinar, which is partially supported by de.NBI, the German Network for Bioinformatics Infrastructure, we will guide you through the complete journey of a data scientist: from training and selecting the best machine learning model for your data to putting your model into production and creating a simple web application.
For this, we will demonstrate a use case of bioactivity prediction.
- Train and optimize four different machine learning methods (Naive Bayes, Logistic Regression, Random Forest, XGBoost)
- Identify the best model to predict the activity of a compound on a particular biological target
- Use KNIME’s new integrated deployment functionality to automatically deploy the best model
- Create a simple web application that uses the deployed model to predict the activity of new compounds
Got questions about integrated deployment or bioactivity prediction? Post them here!
I'd like to read more about the topic of Integrated Deployment. Where can I find more information?
Where can I find the recording of the webinar?
We’ve got one here on KNIME TV on Youtube.
Where can I find the slides from the presentation at the webinar?
I'd like to try out the workflow that was demonstrated. Where can I find it?
Download the workflow, Model Selection with Integrated Deployment, from the KNIME Hub here.
Is it possible to convert PNG to SMILES in KNIME?
There is currently no designated node, however there are two ways to do it:
- The code embeds metadata in the PNG so that you directly have the SMILES. The RDKit will have this functionality in the upcoming version. If you want to try it out immediately, you can but you have to use the Python Scripting nodes with the 2020.09 RDKit version.
- The code converts the image itseld into SMILES. There is software to do this, including open source software, e.g. https://github.com/ncats/molvec