Hi there! I’m here to share some articles I read recently. They mainly focus on protein sequence-based functional annotations and protein interaction analysis or prediction. Although not all of them are the latest, the articles are no older than one year. I wish I could study from them and apply their methods in my future work.
The research presented a sensitive yeast two-hybrid-based pipeline for positive selection of interaction-disruptive variants by developing Int-Seq, a scalable method that can dissect binary PPI at individual amino-acid resolution.
BBSome interaction Network was drawn.
It seems that this method can identify protein interaction change when an amino acid altered.
I think the most interesting part of this job is that it can characterize residues involved in maintaining interaction.
Didn’t find the figures valuable.
This review introduced something I’m familiar with. Boring. But it is a nice collection. It’s worth reading when I want to search for useful tools and algorithms.
Sequence-based prediction of protein-protein interactions using weighted sparse representation model combined with global encoding
A novel prediction model for PPI was developed by combining weighted sparse representation based classifier and global encoding representation of proteins. The figures are convincing but can be improved. It’s a standard research article presenting a method. Not exciting. This article included comparison with SVM and other methods, and pointed out the weakness of methods based on homology and structure.
DeepGO: predicting protein functions from sequence and interactions using a deep ontology-aware classifier
I should have realized the exact name of this tool before starting reading. The article represented a novel method with deep learning to predict protein GO annotation based on sequence. The web server is available here.
Just put this article down, take a deep breath and switch to your own job.
That’s all about it.