A Survey of Computational Intelligence Techniques in Protein Function Prediction

Tiwari, Arvind Kumar; Srivastava, Rajeev

doi:10.1155/2014/845479

Cited by 40 publications

(18 citation statements)

References 152 publications

(173 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Numerous computational methods were developed to predict whether a given protein binds NA. Some rely on overall sequence or structure homology to other NABPs, while others look for homology through shorter sequence signatures or similarity in traits such as amino acid composition41314. Other methods do not try to predict whether a protein binds NA.…”

mentioning

confidence: 99%

De-novo protein function prediction using DNA binding and RNA binding proteins as a test case

et al. 2016

View full text Add to dashboard Cite

Of the currently identified protein sequences, 99.6% have never been observed in the laboratory as proteins and their molecular function has not been established experimentally. Predicting the function of such proteins relies mostly on annotated homologs. However, this has resulted in some erroneous annotations, and many proteins have no annotated homologs. Here we propose a de-novo function prediction approach based on identifying biophysical features that underlie function. Using our approach, we discover DNA and RNA binding proteins that cannot be identified based on homology and validate these predictions experimentally. For example, FGF14, which belongs to a family of secreted growth factors was predicted to bind DNA. We verify this experimentally and also show that FGF14 is localized to the nucleus. Mutating the predicted binding site on FGF14 abrogated DNA binding. These results demonstrate the feasibility of automated de-novo function prediction based on identifying function-related biophysical features.

show abstract

mentioning

confidence: 99%

De-novo protein function prediction using DNA binding and RNA binding proteins as a test case

et al. 2016

View full text Add to dashboard Cite

show abstract

“…It also works with smaller dataset but on them, it can be much stronger and powerful in building models. It uses a subset of training points in the decision function (called support vectors), so it is also memory efficient [31][32][33][34][35]. (vi) Bayesian: It is a graph-based classification technique.…”

Section: Methods Classification Of Protein In Enzyme Classmentioning

confidence: 99%

Rat Protein’s Enzyme Class Classification Using Machine Learning

Gupta,

Bihari,

Tripathi

2019

IJEAT

View full text Add to dashboard Cite

In the current era, bioinformatics has been an emerging research area in the context of protein enzyme classification from the unknown protein data. In bioinformatics, the prime goal is to manipulate the protein data and develop a computational technique to classify and predict the appropriate features for function predictions. In this context, several machine learning and statistical technique have been designed for classification of data. The classification of protein data is one the challenging task and generally the classification of protein data has been done on human protein data. In this article, we have considered rat enzyme class for classification and predictions. Here we have used like CRT, CHAID, C5.0, NEURAL, SVM, and Bayesian for classification of protein data and to measure the performance of the model, the accuracy, specificity, sensitivity, precision, recall, f-measures and MCC have been used. The experimental result highlights that the some of the protein data are imbalance that affects the performance. In this experiment, the Lyases, Isomerases and Ligases class of data are imbalanced and affect the performance of the models. The experimental results highlight that the C5.0 gives 91.5% accuracy and takes only 4 second for computation and can be used for protein classification and prediction of protein data.

show abstract

“…In this technique the learning process is based on adjustment of weight between connection of neurons and the output of the model is depends on the activation function [30]. (vi) SVM: It is one of the most influenced classification techniques based on statistical learning for classifications and prediction of data [31][32][33][34][35]. It deals with wide variety of classification problems including the non-linearly high dimensional problem.…”

Section: Classification Of Protein In Enzyme Classmentioning

confidence: 99%

Human Protein Sequence Classification using Machine Learning and Statistical Classification Techniques

Gupta¹,

Bihari²,

Tripathi³

2019

IJRTE

View full text Add to dashboard Cite

 Abstract: In the field of computational biology, to gauge the meaningful and accurate feature for protein function predications, either the profile-based protein data or sequence-based data has been used. As we know that the prediction of enzyme class from an unknown protein is most interacted research in the current era. In this context, machine learning and statistical classification technique has been used. In this article, we have use six different machine learning and statistical classification technique such as CRT, QUEST, CHAID, C5.0, ANN and SVM for classification of 4314 number of human protein sequence data. These data are extracted form UniprotKB databank with the help of PROFEAT server. The extracted data are categorized in seven different classes. To manipulate the high dimensional protein sequence data with some missing value, the SPSS has been used for classification and estimation of the performance of classification technique. The experimental results highlight that the class C4, C5, C6 and C7 data are imbalanced that affect the overall performance of classification technique. This article provides an extensive comparative analysis of different classification technique on sequence-based protein data. The experimental analysis highlights that the SVM and C5.0 classification technique gives better result than others and can be used for protein classification and predictions.

show abstract

A Survey of Computational Intelligence Techniques in Protein Function Prediction

Cited by 40 publications

References 152 publications

De-novo protein function prediction using DNA binding and RNA binding proteins as a test case

De-novo protein function prediction using DNA binding and RNA binding proteins as a test case

Rat Protein’s Enzyme Class Classification Using Machine Learning

Human Protein Sequence Classification using Machine Learning and Statistical Classification Techniques

Contact Info

Product

Resources

About