SProtP: A Web Server to Recognize Those Short-Lived
Proteins Based on Sequence-Derived Features in Human
Cells
Xiaofeng Song
1
, Tao Zhou
2
, Hao Jia
1
, Xuejiang Guo
2
*, Xiaobai Zhang
1
, Ping Han
3
*, Jiahao Sha
2
1 Department of Biomedical Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing, China, 2 State Key Laboratory of Reproductive Medicine,
Department of Histology and Embryology, Nanjing Medical University, Nanjing, China, 3 Department of Gynecology and Obstetrics, The First Affiliated Hospital with
Nanjing Medical University, Nanjing, China
Abstract
Protein turnover metabolism plays important roles in cell cycle progression, signal transduction, and differentiation. Those
proteins with short half-lives are involved in various regulatory processes. To better understand the regulation of cell
process, it is important to study the key sequence-derived factors affecting short-lived protein degradation. Until now, most
of protein half-lives are still unknown due to the difficulties of traditional experimental methods in measuring protein half-
lives in human cells. To investigate the molecular determinants that affect short-lived proteins, a computational method
was proposed in this work to recognize short-lived proteins based on sequence-derived features in human cells. In this
study, we have systematically analyzed many features that perhaps correlated with short-lived protein degradation. It is
found that a large fraction of proteins with signal peptides and transmembrane regions in human cells are of short half-lives.
We have constructed an SVM-based classifier to recognize short-lived proteins, due to the fact that short-lived proteins play
pivotal roles in the control of various cellular processes. By employing the SVM model on human dataset, we achieved
80.8% average sensitivity and 79.8% average specificity, respectively, on ten testing dataset (TE1-TE10). We also obtained
89.9%, 99% and 83.9% of average accuracy on an independent validation datasets iTE1, iTE2 and iTE3 respectively. The
approach proposed in this paper provides a valuable alternative for recognizing the short-lived proteins in human cells, and
is more accurate than the traditional N-end rule. Furthermore, the web server SProtP (http://reprod.njmu.edu.cn/sprotp) has
been developed and is freely available for users.
Citation: Song X, Zhou T, Jia H, Guo X, Zhang X, et al. (2011) SProtP: A Web Server to Recognize Those Short-Lived Proteins Based on Sequence-Derived Features
in Human Cells. PLoS ONE 6(11): e27836. doi:10.1371/journal.pone.0027836
Editor: Emanuele Buratti, International Centre for Genetic Engineering and Biotechnology, Italy
Received July 4, 2011; Accepted October 26, 2011; Published November 16, 2011
Copyright: ß 2011 Song et al. This is an open-access article distributed under the terms o f the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The study was supported by grants from National Key Basic Research Program Grant (2011CB944304), National Natural Science Foundation of China
(No.61171191), Specialized Research Fund for the Doctoral Program of Higher Education (20103234120002), and Natural Science Foundation of Jiangsu Province
in China (BK2010500). The funders had no role in study desig n, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: guo_xuejiang@njmu.edu.cn (XG); hanping200701@163.com (PH)
Introduction
Proteins are the chief actors within the cell. All proteins in
mammal cells are continually being degraded and replaced. Some
cytosolic enzymes have half-lives as short as 10 minutes, whereas
others last for days. The fluctuations in their expression are
fundamental for metabolism, cell cycle control and communica-
tion between a cell and environment. The cell’s proteolytic
machinery must be highly selective and tightly regulated, since the
accelerated destruction of an essential protein or the failure to
degrade a short-lived regulatory protein could drastically alter cell
function [1].
The continual destruction of cell proteins may appear to be
wasteful, but it serves several important homeostatic functions [2].
The rapid removal of rate-limiting enzymes and regulatory
proteins is essential for the control of growth and metabolism.
For example, the progression of cells through the mitotic cycle
requires the programmed destruction of the critical regulatory
proteins called cyclins. The rapid degradation of specific proteins
permits an adaptation to new physiologic conditions. And the
tumor suppressor protein p53 is a short-lived protein that is
maintained at low, often undetectable, levels in normal cells [3]. If
p53 was stabilized in response to an activating signal, such as DNA
damage, its expression level will rise rapidly and inhibit cell
growth.
The study of factors affecting those short-lived proteins has
begun in the last century. The amino acid sequence composition
was shown to be closely related to protein half-lives. In 1986,
Rogers et al. proposed the PEST hypothesis that proteins with
PEST sequence tend to undergo rapid intracellular degradation
[4]. In the same year, Bachmair et al. proposed the N-end rule
that certain terminal amino acid yielded proteins with very short
half-lives whereas others rendered the protein very stable [5] [14].
Other protein sequence motifs such as D-box and KEN box were
also found to be important for protein stability regulation [6–8].
Most of above studies for protein degradation were previously
performed on only a few or individual proteins due to the technical
difficulty in global profiling. While in 2006, Archana et al.
performed profiling of over 3,000 protein stability in yeast [9].
Follow-up bioinformatics analysis of this data showed that among
all the physical and sequence features of proteins considered, the
most significant feature that correlates with intracellular degrada-
PLoS ONE | www.plosone.org 1 November 2011 | Volume 6 | Issue 11 | e27836