Review: prediction of in vivo fates of proteins in the era of genomics and proteomics

Even after a nascent protein emerges from the ribosome, its fate is still controlled by its own amino acid sequence information. Namely, it may be co-/posttranslationally modified (e.g., phosphorylated, N-/O-glycosylated, and lipidated); it may be inserted into the membrane, translocated to an organelle, or secreted to the outside milieu; it may be processed for maturation or selective degradation; finally, its fragment may be presented on the cell surface as an antigen. Here, prediction methods of such protein fates from their amino acid sequences are reviewed. In many cases, artificial neural network techniques have been effectively used. The prediction of in vivo fates of proteins will be useful for characterizing newly identified candidate genes in a genome or for interpreting multiple spots in proteome analyses.