Welcome to LabDAO | LabDAO Documentation
Equibind is a very fast, machine learning-based docking tool. The model is less accurate than baseline methods, but orders of magnitude faster.
./plex create -t tools/equibind.json -i ~/downloads/plexbindingdata --autoRun=true
Diffdock is a machine learning-based docking tool. Diffdock is reportedly faster and more accurate than existing baseline tools.
./plex create -t tools/diffdock.json -i ~/downloads/plexbindingdata --autoRun=true
Gnina is a sampling and machine learning-based docking tool. Gnina is an implementation of Smina, which itself is a fork of Vina. These tools are considered the current open source baseline.
Colabfold is an implementation of Alphafold that uses a multiple sequence alignment (MSA) Server, MMSeq2, instead of a local database to make using Alphafold more lightweight. The "mini" configuration runs a shallow MSA, performs one recycling and uses available templates to make a prediction. This is best used if you want to predict a protein structure very fast.
./plex create -t tools/colabfold-mini.json -i testdata/folding --autoRun=true
The "standard" configuration runs a full MSA, performs three recycling rounds and uses available templates to make a prediction. It runs this prediction 5 times with different randomness seeds. This is best used if you want to predict a state of the art structure and draw from a distribution of potential conformational substates.
./plex create -t tools/colabfold-standard.json -i testdata/folding --autoRun=true
The "large" configuration runs just like the standard configuration, but includes a GPU-accelerated relaxation step using Amber. It returns 25 predictions. This is best used when a lot of ressources are available and you want to predict a state of the art structure while drawing from a larger distribution of potential conformational substates.