Graphical Representation and Analysis of Biological Sequences

Graphical representation and analysis of biological sequences encompasses a suite of alignment-free methodologies that convert nucleotide or amino acid strings into visual or numerical forms, facilitating rapid comparison, classification and interpretation. Early approaches mapped bases to curves or matrices, enabling intuitive detection of motifs, repeats and global compositional trends without the need for pairwise alignment. Subsequent innovations introduced multidimensional embeddings—such as two- and three-dimensional plots, spiral curves and energy-based matrices—that capture both local order and long-range correlations. These representations serve as fingerprints for genomes, transcripts and proteomes, supporting phylogenetic inference, mutation detection and functional annotation. By abstracting biochemical properties into spatial coordinates or numerical descriptors, researchers have achieved scalable analyses of large-scale sequencing data, accelerated machine-learning pipelines and improved robustness to sequence rearrangements. The integration of statistical, algebraic and graphical features continues to expand the toolkit for exploring sequence diversity and elucidating evolutionary relationships across all domains of life.

Research from Nature Portfolio

Recent studies have refined energy-based graphical frameworks to enhance protein comparison. One approach constructs a position-feature energy matrix that integrates physicochemical attributes of amino acids with sequence order information, yielding characteristic vectors that improve clustering and classification over traditional alignment. Another work employs fuzzy integral techniques within an alignment-free context, estimating Markov parameters for amino acid pair frequencies and aggregating them via fuzzy measures to achieve superior clustering performance in benchmark datasets. A third development combines global and local position information of DNA sequences by projecting nucleotides onto a Fermat spiral and assigning mass based on neighbourhood relationships; the resulting inertia moments provide compact numerical descriptions that distinguish species and support evolutionary analyses.

Graphical Representation and Analysis of Biological Sequences publication trend

The graph below shows the total number of articles in graphical representation and analysis of biological sequences across all publications each year (not limited to Nature Index journals).

Technical terms

Graphical representation: Visual or spatial mapping of biological sequences to capture compositional and structural features.

Chaos Game Representation (CGR): Fractal-based method that maps sequence k-mers to coordinates within a unit square, revealing compositional patterns.

Position-Feature Energy Matrix: Descriptor matrix quantifying sequence order through energy calculations on physicochemical properties of residues.

Fuzzy integral: Aggregation operator that combines similarity measures, accounting for interactions among sequence features in alignment-free comparisons.

k-mer: Subsequence of length k extracted from a biological sequence, commonly used to summarise local composition.

References

1. Protein Sequence Comparison Based on Physicochemical Properties and the Position-Feature Energy Matrix. Scientific Reports (2017).
2. Alignment-free similarity analysis for protein sequences based on fuzzy integral. Scientific Reports (2019).
3. One novel representation of DNA sequence based on the global and local position information. Scientific Reports (2018).
4. Digitizing DNA Sequences Using Multiset-Based Nucleotide Frequencies for Machine Learning-Based Mutation Detection. Decision Making Applications in Management and Engineering (2024).
5. aaHash: recursive amino acid sequence hashing. Bioinformatics Advances (2023).
6. FEGS: a novel feature extraction model for protein sequences and its applications. BMC Bioinformatics (2021).