1 - 5 of 5 results

SLIPPER / SeLf-Interacting Protein PrEdictoR

Predicts proteome-wide of self-interacting proteins. SLIPPER uses various related annotation information such as interaction partners in the protein interaction network (PIN), domains, and evolutionary rate. It depends on the currently available related biological data. The tool can return probability scores and abundant annotations for proteins submitted by users. It is based on logistic regression and provides features like model organism self-interacting protein.

WELM-LAG-SIPs / Weighed-Extreme Learning Machine - Local Average Group- Self-Interactions Proteins

Predict self-interactions proteins (SIPs) using protein sequence information. WELM-LAG-SIPs’ method is a computational approach that combines a feature extraction method called Local Average Group (LAG) and a classifier called the Weighed-Extreme Learning Machine (WELM). The software is able to extract the hidden key information beyond the sequence itself. It was applied on human and yeast datasets.


Predicts self-interacting proteins (SIPs) using the information of protein sequences. PSPEL is a matrix-based method that (1) transforms all protein sequences into position-specific-scoring-matrix (PSSM) representation based on protein sequences, (2) uses the low-rank approximation (LRA) descriptor to capture the useful information for SIPs prediction from each protein PSSM, (3) generates a final protein feature vectors and (4) finally uses the machine learning technology to predict SIPs.

SPAR / Self-interacting Protein Analysis serveR

Predicts self-interacting proteins (SIPs). SPAR is based on an improved sequence-encoding scheme that exploits the fine-grained domain–domain interaction (DDI) information from the 3did database. It was tested on the human SIP dataset and also compared with other popular feature-encoding methods commonly used for protein-protein interaction (PPI) prediction. The tool facilitates the high-throughput prediction analysis of protein self-interactions.