Sequencewide

Sequencewide is an adjective used in biology, bioinformatics, and related fields to describe analyses, models, or metrics that treat an entire sequence as the unit of observation rather than focusing on individual positions or specific motifs. In genomics, proteomics, and transcriptomics, sequencewide approaches aim to capture global properties and dependencies across the full length of a sequence, whether it is a DNA, RNA, or protein sequence.

Common sequencewide concepts include sequencewide conservation scores, which summarize how invariant a whole sequence is across a family; sequencewide mutational effect predictions, which assess the potential impact of substitutions at all positions within the sequence; and sequencewide motif scanning, which evaluates motif presence in the context of the entire sequence. In machine learning and data analysis, sequencewide methods may involve sequencewide losses that aggregate error over all positions, or sequencewide attention and modeling that consider long-range relationships across the full sequence.

The term is often contrasted with sitewise, positionwise, or local analyses, which report metrics or predictions for individual positions rather than the entire sequence. Sequencewide approaches can provide a global view useful for understanding overarching sequence properties, guiding design and benchmarking, or informing models that need a holistic view of sequence structure. Despite its utility, sequencewide is used informally, and in some contexts researchers may prefer sequence-level or whole-sequence terminology.