Nfeature: A platform for computing features of nucleotide sequences

Abstract

In the past few decades, public repositories on nucleotides have increased with exponential rates. This pose a major challenge to researchers to predict the structure and function of nucleotide sequences. In order to annotate function of nucleotide sequences it is important to compute features/attributes for predicting function of these sequences using machine learning techniques. In last two decades, several software/platforms have been developed to elicit a wide range of features for nucleotide sequences. In order to complement the existing methods, here we present a platform named Nfeature developed for computing wide range of features of DNA and RNA sequences. It comprises of three major modules namely Composition, Correlation, and Binary profiles. Composition module allow to compute different type of compositions that includes mono-/di-tri-nucleotide composition, reverse complement composition, pseudo composition. Correlation module allow to compute various type of correlations that includes auto-correlation, cross-correlation, pseudo-correlation. Similarly, binary profile is developed for computing binary profile based on nucleotides, di-nucleotides, di-/tri-nucleotide properties. Nfeature also allow to compute entropy of sequences, repeats in sequences and distribution of nucleotides in sequences. In addition to compute feature in whole sequence, it also allows to compute features from part of sequence like split-composition, N-terminal, C-terminal. In a nutshell, Nfeature amalgamates existing features as well as number of novel features like nucleotide repeat index, distance distribution, entropy, binary profile, and properties. This tool computes a total of 29217 and 14385 features for DNA and RNA sequence, respectively. In order to provide, a highly efficient and user-friendly tool, we have developed a standalone package and web-based platform (https://webs.iiitd.edu.in/raghava/nfeature).

A hybrid approach for predicting multi-label subcellular localization of mRNA at genome scale

“…In order to train our model, we are required to generate features or descriptors corresponding to each mRNA. For the aforementioned purpose, we have used the tool ‘Nfeature’ [20] [https://doi.org/10.1101/2021.12.14.472723], which can generate hundreds of features for a single mRNA sequence. These are the two feature classes which were used for training the models: Composition of DNA/RNA for k-mer (CDK): k-mers of length 3 were generated by Nfeature and the frequency of each k-mer was used as a feature for training the ML model.…”

Sigma70Pred: A highly accurate method for predicting sigma70 promoter in Escherichia coli K-12 strains

“…We have generated a wide range of features like Position-Specific Tri-Nucleotide Propensity (PSTNPP), Electron-Ion Interaction Pseudopotentials of trinucleotide (EIIIP; He et al, 2018 ), dimer count, trimer count, motif counts, GC and AT skew ( Rahman et al, 2019a ), Dinucleotide Auto-Correlation (DAC), Dinucleotide Cross-Correlation (DCC), Dinucleotide Auto Cross-Correlation (DACC; Friedel et al, 2009 ), Moran Auto-Correlation (MAC), Normalized Moreau-Broto Auto-Correlation (NMBAC; Chen et al, 2015 ), and Parallel Correlation Pseudo Tri-Nucleotide Composition (PC_PTNC; Liu et al, 2014 ), which resulted in 8465 features. The aforementioned features were calculated using Nfeature webserver ( Mathur et al, 2021 ) available at https://webs.iiitd.edu.in/raghava/nfeature/ . Then, we have used the Min-Max scaler from the scikit-learn library ( Pedregosa et al, 2011 ) to scale down the values of the features, we have constructed.…”

A random forest model for predicting exosomal proteins using evolutionary information and motifs

“…A number of feature encoding techniques have been used in previous studies [27][28][29][30]. We used a standalone tool called Pfeature to compute numerous features for the proteins, including evolutionary information-based features and composition-based features [31].…”