how should I approach gene expression datasets
I'm a comp sci undergrad student, and currently doing my thesis. So my supervisor has assigned bioinformatics as my domain and said to look for something regarding gene expression. My initial thought was to involve graph-based (deep learning and general) approaches, catering to good research work. However, I'm unable to find labelled datasets or unable to structure a proper dataset from the Gene Expression Omnibus. How do I curate the dataset? And particularly for gene expression, what are the things I need to learn thoroughly?