Here are some of the projects I've been working on, organized in categories:
Legend (for categories):
Bio = Biophysics
ML = Machine Learning
DB = Database (tipically involving SQL)
[Bio] EDES tutorial.
Here you will find a tutorial for the protocol the "Enhanced Sampling of Binding and Pocket Volume and Shape" (EDES) developed during my Ph.D. (link to the publication). In the tutorial, we will exploit enhanced-sampling molecular dynamics, namely metadynamics, to improve the sampling of bound-like (holo) conformations of a generic protein starting from its unbound (apo) one.
[ML/Bio] K-means clustering of protein conformations.
Here a tutorial about the usage of K-means clusteringon protein conformations used for teaching. However, the whole code can be used with any dataset and it is not protein-specific. The code is based on the Scikit-learn library, and it also implements: (i) feature normalization; (ii) identifying cluster representatives from centroids; (iii) the elbow and silhouette methods for the choice of the optimal number of clusters. The folder also contains a tcl script to extract some of the features employed in the tutorial from a molecular dynamics trajectory using the VMD software.
[ML] Sentiment Analysis.
Here a simple implementation of the sentiment analysis. Given a sentence, the model predicts if it conveys a positive, negative or neutral feeling. The code is based on the "Natural Language ToolKit" (NLTK) python library. A logistic regression model has been used, minimizing the cost function with the gradient descent method. Both logistic model and the gradient descent method have been implemented from scratch.