Fractal and DNA sequence analysis

The paper shows that nullomers in genomes arise from statistical properties rather than evolutionary pressures or lethal effects.

UniWave-2 improves nucleic acid sequence analysis by integrating hydrophobicity, enhancing waveform resolution with Fourier and wavelet transforms, and using a lightweight GRU model for efficient feature extraction.

The paper maps long-range correlations in the complete telomere-to-telomere human genome, revealing heterogeneous genomic structure and chromosome-specific scaling.

This paper introduces an alignment-free method using information-theoretic features to detect gene fusions, achieving strong performance in classification and partner prediction.

The paper analyzes whole genome sequences to identify non-random patterns in restriction site distributions, showing deviations from random sequences and effects of GC content and repetitive elements.

The paper demonstrates that the human genome encodes a deterministic computational kernel with four properties, validated through analysis of proteome and chromosome data.

This paper introduces a quaternion Fourier transform framework for DNA sequence analysis, demonstrating its ability to detect structural periodicities and spectral signatures across multiple species.

The paper introduces information-theoretic descriptors to analyze protein sequences as statistical languages, showing that real proteomes exhibit structured dependencies beyond simple compositional models.

dna-parser encodes DNA and RNA sequences into numerical features efficiently using Rust and Python, offering fast and parallel processing.

UniWave converts nucleotide sequences into continuous waveforms for improved genomic analysis, outperforming traditional methods in multiple tasks.

This study evaluates neural networks trained with natural vectors incorporating covariance for predicting plant microRNA precursors, showing improved performance in feature prediction and classification.

This study shows that high dimensional silent patterns in viral coding regions encode regulatory codes related to viral tropism using a novel computational measure.

The paper identifies over 19 million inverted repeats in 14,000 viral genomes using BLMT, highlighting their prevalence and potential biological significance.

This paper introduces a DNA sequence imaging method for splice site prediction, showing that DFCP outperforms FCGR using deep learning on Arabidopsis and human data.

This paper introduces SVD-FCGR, an alignment-free phylogenetic analysis method using SVD on FCGR, showing improved resolution and scalability over MEGA for viral datasets.

This paper converts protein sequences into music using NMR data, assigning notes based on amino acid frequencies and relaxation periods, and applies set theory to analyze homologous beta-globin sequences.

TIPs-VF provides a variable-length DNA sequence encoding framework that improves representation of genetic sequences for machine learning by incorporating sequence, length, and positional awareness.

This study prioritizes lung cancer candidate genes using moment of inertia tensor analysis of amino acid physicochemical properties, identifying four genes linked to poor survival.

The paper introduces a Python package for DNA sequence encoding using Frequency Chaos Game Representation (FCGR), showing improved accuracy and efficiency over an R-based library for DNA sequences.

The paper introduces a method to detect whether viruses are double or single stranded by analyzing dinucleotide selective profiles in viromes.

The paper analyzes the Chaos Game Representation of genomic sequences to assess ergodicity and finds that CGRs are non-ergodic due to distribution properties of nucleotide sequences.

This paper develops a fast feature extraction method for SARS-CoV-2 spike sequences using amino acid physicochemical properties, achieving accurate clustering with reduced runtime compared to conventional methods.

The paper describes the application of the Square Wave Method to analyze human DNA sequences and presents preliminary data supporting its potential functional relevance.

The paper identifies 210 mirror repeat sequences in the human insulin gene using an in silico approach called FPCB.

ortho_seqs quantifies sequence-phenotype interactions using tensor-based orthogonal polynomials, demonstrating its application to synthetic antibody sequences and other datasets.

The paper develops a sequence-based chaos game representation combined with deep learning for peptide/protein classification, showing superior performance in hemolytic/non-hemolytic peptide and HIV protease mutant classification.

The paper analyzes ATG-walks of Omicron SARS-CoV-2 RNA sequences to compare their statistical characteristics with a reference RNA, finding small deviations and resistance to forming a new virus family.

Nfeature computes a large number of features for DNA and RNA sequences, providing a standalone package and web platform for researchers.

This paper applies stacked sparse autoencoders to classify SARS-CoV-2 genomes using k-mer image representations, achieving good performance in taxonomic classification.

DiviSSR uses a mathematical property of 2-bit DNA representations to quickly and accurately identify tandem repeats in genomic data.