With the development of high-throughput sequencing techniques, increasing amounts of protein data have become available. In these proteins, some display special functions. The knowledge about these proteins could provide an opportunity to explore new targets for disease treatment.

Thus, it is urgent for us to develop computational methods to study and analyze these special functional proteins. As such, more and more scholars have focused on this topic. Some computational methods have been developed for the prediction of protein subcellular localization and the identification of post-translational modification sites. However, the function of many proteins has still not been annotated. As of July 17th 2020, the Uniprot database contains 184,998,855 proteins. However, the database provides the annotation information of only 562,755 proteins. Although some sequence similar algorithms could provide some useful information for these non-annotated proteins, the homologue for many proteins cannot be found in the database. Thus, these similarity-based computational tools cannot give suitable predicted annotation on these proteins. Therefore, machine learning-based methods have increasingly attracted attention.

Due to the rapid development of this field, this Special Issue will mainly focus on the development of machine learning methods to recognize proteins with special functions. We invite authors to contribute original research or review articles in this field.

Potential topics include but are not limited to the following:

• Development of sequence feature extraction methods in special functional proteins
• Applying new mathematical methods to formulate special functional protein samples
• Novel non-sequence feature extraction in special functional protein description
• Feature fusion in ion channel protein identification
• Computational methods in cyclin protein prediction
• Toxin protein prediction and analysis using machine learning methods
• Receptor protein identification using computational methods
• Developing new tools for hormone-related protein identification
• Special enzyme identification using machine learning methods
• Immunoglobulins recognition