CrossLink

CrossLink visualizes pairwise sequence similarity among RNA sequences and constructs attribute-annotated networks to analyze microRNA-target and other RNA interactions.

Key Features:

• Integration with BLAST and Vmatch: Uses BLAST and Vmatch to detect pairwise sequence similarity and matches among nucleotide sequences.
• RNAhybrid integration: Uses RNAhybrid to determine antisense and hybridization relationships relevant to microRNA interactions.
• Initial parametrization: Performs an initial parametrization phase to configure match detection parameters.
• Network construction: Represents each nucleotide sequence as a node and each match or set of matches as a link in a network.
• Attribute annotation: Computes and associates attributes with links such as E-values, match lengths, and proportions of identical nucleotides, and provides detailed alignments.
• Attribute histograms: Summarizes distributions of match attributes using histograms for dataset-wide overview.
• Threshold-based filtering: Allows highlighting or suppression of matches by adjusting thresholds based on combinations of link attributes.
• Network layout algorithms: Applies network layout operations including spring-embedding algorithms to arrange network visualizations.

Scientific Applications:

• microRNA-target interaction analysis: Identifies and characterizes antisense and hybridization relationships between microRNAs and target sequences using Vmatch and RNAhybrid.
• RNA sequence similarity exploration: Explores complex pairwise similarity relationships among RNA sequences via network representations.
• Dataset-wide attribute analysis: Examines distributions of E-values, match lengths, and nucleotide identity proportions across datasets using histograms and thresholding.

Methodology:

Performs an initial parametrization, integrates BLAST, Vmatch, and RNAhybrid to detect matches and hybridization, constructs networks with nodes as sequences and links as matches, computes link attributes (E-values, match lengths, proportions of identical nucleotides) and detailed alignments, summarizes attributes in histograms, applies threshold-based filtering, and uses spring-embedding algorithms for network layout.