Nacho Garcia's project group: Artificial intelligence in life sciences
The rate at which life sciences generate extensive and complex data highly exceeds our present analytical capacity and the current tools are still unable to fully identify the meaningful information depending on complex interactions laying in the data.
The use of advanced analytical methods such as deep learning or predictive analytics can help us to decipher and understand the underlying relationships within the data, otherwise intractable.
In this project-group, we are developing these advanced analytical tools to solve complex biological problems in many different fields such as image analysis, drug discovery or advanced clinical data analysis.
Our final goal is to generate these advanced tools and to make them publicly available for the scientific community.
Machine learning-based microscopy analyses:
In this project, we are developing a set of tools based on artificial neural networks (NN) and convolutional neural networks (CNN) to process and analyze high throughput microscopy data to automatically identify, classify and predict complex cellular processes.
Computational drug discovery:
We are investigating novel models to perform advanced structure-activity relationship studies that can be used to perform virtual screenings against massive drug databases.
We are also integrating large drug-protein interaction datasets to create models able to find novel drugs against disease-specific targets.
AI-based synthetic biology:
We are exploring the use of CNN ensembles to find molecular signatures in the genome. Our final goal is to create a model capable of generating these molecular signatures “de novo”.
Quantitative clinical outcome assessment:
We are developing machine learning-based models to correlate patterns in early clinical data with the progression of the disease to gain a deeper understanding of the disease in order to offer better treatment for the patients.