Related Experiment Video
Updated: May 12, 2026

Enriching Subcellular Proteins in Leptospira Using a Triton X-114-Based Fractionation Approach
Published on: August 8, 2025
SCL-Epred: a generalised de novo eukaryotic protein subcellular localisation predictor
Catherine Mooney1, Amélie Cessieux, Denis C Shields
1Complex and Adaptive Systems Laboratory, Conway Institute of Biomolecular and Biomedical Science, School of Medicine and Medical Science, University College Dublin, Ireland. catherine.mooney@ucd.ie
Accurate prediction of eukaryotic protein subcellular localization is crucial for understanding protein function and drug targetability. SCL-Epred, a novel N-to-1 neural network predictor, efficiently classifies proteins into secreted, membrane, or non-secreted/non-membrane categories.
Area of Science:
- Bioinformatics
- Computational Biology
- Molecular Biology
Background:
- Determining protein subcellular localization is vital for understanding protein function, interactions, and drug target potential.
- Experimental methods for localization are costly, time-consuming, and prone to error.
- The post-genomic era necessitates rapid and accurate computational tools for analyzing vast protein sequence data.
Purpose of the Study:
- To develop a general eukaryotic subcellular localization predictor (SCL-Epred) for classifying proteins into three key functional categories.
- To improve the prediction accuracy of protein subcellular localization, aiding in drug target identification.
- To provide a publicly accessible tool for researchers studying eukaryotic protein localization.
Main Methods:
- Development of SCL-Epred, a predictor utilizing an N-to-1 neural network algorithm.
- Training the algorithm on large, non-redundant sets of eukaryotic protein sequences.
- Performance evaluation using tenfold cross-validation and comparison with existing predictors on independent test sets.
Main Results:
- SCL-Epred achieved a Q score of 86% and a generalized correlation of 0.75 on training data.
- The predictor demonstrated high accuracy in classifying proteins into secreted, membrane, and other categories.
- A consensus predictor combining SCL-Epred and LocTree2 outperformed other widely used methods on an independent test set.
Conclusions:
- SCL-Epred offers a fast and accurate method for predicting eukaryotic protein subcellular localization.
- The predictor is particularly valuable for identifying potential drug targets based on protein location.
- SCL-Epred is publicly available, facilitating broader research in eukaryotic cell biology and drug discovery.
Related Concept Videos
Signal Sequences and Sorting Receptors
Regulation of Nuclear Protein Sorting
Nuclear Protein Sorting
Proteins targeted to the nucleus carry nuclear localization signals or NLS recognized by import receptors in the cytosol. Similarly, proteins with nuclear export signals are recognized by export receptors. Import and export receptors are...
Directing Proteins to the Rough Endoplasmic Reticulum
Overview of Protein Sorting and Transport
Protein sorting can be of two types: signal-based sorting and vesicle-based trafficking. In signal-based sorting, specific amino acid sequences called sorting signals target proteins to the proper location inside the cell either via gated transport or by protein translocation. In gated transport, folded...
Post-translational Translocation of Proteins to the RER
Targeting proteins to the ER
Hsp40 and Hsp70 chaperone molecules bind the translated proteins in the cytosol to prevent their folding. The chaperone binding helps to keep the signal...

