Machine Learning

	Title	Description	Input Data
	MLflow Example Model (Wine Preference)	This is an example Capsule to show the functionality of the MLFLow Orchestrator Capsule. It generates a model, logs parameters and results into the mlflow database and then syncs the results to S3.	.csv file containing wine quality metrics
	MLflow Orchestrator Capsule	This is a central MLflow "Orchestrator" Capsule that can be used to interactively view the results from multiple machine learning Capsules using the same MLflow database	None
	MLFlow Example Model (Diabetes Experiment)	This is an example Capsule to show the functionality of the MLFLow Orchestrator Capsule. It generates a model, logs parameters and results into the mlflow database and then syncs the results to S3.	.csv file containing diabetes metrics for various patients.
	DECIMER 2.0	This Capsule uses DECIMER (Deep Earning for Chemical ImagE Recognition) 2.0 to predict the SMILES (simplified molecular input line entry system) for input chemical images.	.jpg or .png images of chemical compounds
	Img2Mol	This Capsule uses Img2Mol to predict the compound structure for input chemical images. It will output a structure data file and SMILES representation for each input image.	.jpg or .png images of chemical compounds
	Streamlit ColabFold: AlphaFold2 using MMseqs2	ColabFold offers an accelerated prediction of protein structures and complexes by combining the fast homology search of MMseqs2 with AlphaFold2. This Capsule can be accessed through a Streamlit Cloud Workstation, run in a Pipeline or accessed using the App Panel.	.fasta sequences of proteins to predict the structure for.
	CombFold - prepare	This Capsule generates fasta files of protein subunits for downstream processing by AlphaFold. The subunits are then assembled using Streamlit ColabFold: AlphaFold2 using MMseqs2	subunits.json (see README)
	CombFold- Combinatorial Assembly	The CombFold Combinatorial Assembly algorithm assembles subunits from AlphaFold into a single large complex.	subunits.json (see README) pdb files for each subunit.