Applications of CNNs in genomics

Now that you understand how CNNs work in genomics with a simple example, let’s look at some of their applications that are popular in genomics.

DeepBind

TFs are DNA- and RNA binding proteins that play a crucial role in gene regulation. Knowing the binding sites of these DNA- and RNA-binding proteins would help us to develop models and can help identify disease-causing variants. One way to infer the sequence specificities of these proteins is through position weight matrices (PWMs), which can be used to scan the entire genome to identify potential binding sites of these DNA- and RNA-binding proteins. In addition, DNA- and RNA-binding protein specificities are measured by several high-throughput assays such as PBM, SELEX, and the most popular ChIP- and CLIP-seq techniques. Some of the challenges associated with this data are that the raw data comes in several different quantitative forms, the data is often extremely huge, and each data generation...