KeeSeeK

KeeSeeK identifies absent k-mers (nullomers or "neverwords") in reference genomes to enable selection of primer-like sequences for unique labeling and PCR-based localization in molecular biology.

Key Features:

• Identification of Absent Sequences: Detects k-mers that do not occur within one or more reference genomes.
• Primer-like Feature Detection: Identifies absent k-mers with primer-like characteristics that can serve as unique labels for exogenously inserted DNA fragments.
• Distance Calculation and Selection: Calculates mismatch distance from the reference genome and selects sequences with maximal mismatch distance to minimize nonspecific annealing.
• Filtering for PCR Primer Suitability: Applies filters to eliminate candidate sequences that are unsuitable as PCR primers.
• High-Performance Computing: Implemented in C++ and CUDA to leverage GPGPU environments for efficient, scalable processing of large genomic datasets.

Scientific Applications:

• Unique labeling of exogenous DNA: Provides unique sequence labels for exogenously inserted DNA fragments to enable specific detection.
• PCR-based localization: Enables precise localization of labeled fragments within genomes via PCR.
• Genetic engineering: Supports selection of unique primer-like sequences for use in genetic engineering workflows.
• Diagnostics: Facilitates development of PCR assays that use absent k-mers as specific targets in diagnostic applications.
• Foreign DNA insertion tracking: Assists applications requiring insertion of foreign DNA into genomes by providing distinguishable primer-like sequences.

Methodology:

Analyzes reference genome sequences to identify absent k-mers, computes mismatch distances to select maximally distant sequences, and applies PCR primer-quality filters; implemented in C++ and CUDA to run on GPGPU hardware.