This tool is based on criteria outlined in the N1C VARIANT Guidelines (Cheerie et al. 2025 PMID: 40139194). The guidelines aim to provide a structured framework for analyzing and classifying disease-causing variants based on their potential amenability to ASO therapies. Specifically, the guidelines facilitate identifying variants most likely to benefit from an ASO-based approach, and address the critical first step in evaluating an individual patient for potential ASO development.
Use of HAWK-EYE should be accompanied by a thorough literature review and expert manual evaluation. It is intended to support, not replace, careful variant curation and individualized assessment.
As ASO technologies and the broader therapeutic landscape continue to evolve, these recommendations will be updated accordingly.
Development of the HAWK-EYE web application was supported by the Data Sciences Institute at the University of Toronto, and the Azrieli Precision Child Health Platform.
For the most current version of the guidelines, visit:
N=1 Collaborative - Variant GuidelinesThe exon skipping database in HAWK-EYE is comprised of 72,644 exons from 5,057 protein-coding genes associated with human monogenic diseases (the “Mendeliome”). At present, all exon classifications are based on MANE Select transcripts.
Dispensable exons are coding exons predicted to be inessential for the function or stability of the resulting protein. Skipping dispensable exons produces an in-frame transcript that is expected to generate a truncated but functional protein product. Indispensable exons are coding exons predicted to be essential for the function or stability of the resulting protein. Skipping indispensable exons typically results in frameshifts, premature termination codons, or loss of critical functional domains.
Column Title
Description
Assessment
Classification of exon “skippability” based on the underlying criteria. Exons are classified as either "Dispensable", "Probably Dispensable", "Possibly Dispensable", "Unlikely Dispensable" or "Indispensable".
Gene
NCBI RefSeq official symbol for each gene of interest.
Exon
Exon of interest within the MANE Select transcript.
Transcript ID
RefSeq accession for the MANE Select transcript used for a gene, representing the clinically preferred transcript.
Total Exon Count
The total number of exons within the MANE Select transcript for a given gene.
Strand
Indicates the DNA strand (positive or negative) on which the gene is located.
Chromosome
The chromosome on which the gene is located, using standard human genome nomenclature (e.g., chr1, chrX).
Exon Start
Exon End
Exon Length
The genomic coordinates of the exon (start and end positions) and its length in base pairs. Coordinates are based on the GRCh38 human genome assembly.
Coding Sequence (%)
The percentage of the exon that contributes to the coding sequence (CDS) of the gene.
Exon Frame
Indicates whether skipping the exon maintains the reading frame (In-Frame) or causes a frameshift (Out-of-Frame).
New Junction Codon
The new codon formed at the exon-exon junction when the exon is skipped.
Premature Stop Codon
Indicates whether skipping the exon would introduce a premature stop codon.
First Coding Exon
Indicates whether the exon is the first coding exon in the transcript.
Last Coding Exon
Indicates whether the exon is the last coding exon in the transcript.
LP/P Missense Count
The number of likely pathogenic/pathogenic missense variants from ClinVar located within this exon.
LP/P Nonsense Count
The number of likely pathogenic/pathogenic nonsense variants from ClinVar located within this exon.
LP/P In-Frame Deletions
The number of likely pathogenic/pathogenic in-frame deletions from ClinVar located within this exon.
UniProt ID
The UniProt accession ID for the protein encoded by the gene.
Amino Acid Positions
The positions of amino acids within the protein encoded by the gene in the exon.
Functional Domains
Functional domains within the exon, as annotated in UniProt.
Repeats
Repeats within the exon, as annotated in UniProt.
Other Functional Annotations
Other functional annotations within the exon, as annotated in UniProt, including regions, compositional biases, motifs, zinc fingers, modifications, topological domains, transmembrane domains, coiled coil domains
Post Translational Modifications
Post-translational modifications within the exon, as annotated in UniProt, including signal peptide, transit peptide, crosslinks, disulfide bonds, glycosylation sites, initiator methionine, lipidation sites, peptide, propeptide, active sites, binding sites, DNA binding regions, and other sites.
InterPro Domains
Presence of predicted domains, annotated by InterPro.
The knockdown database in HAWK-EYE contains prioritized common heterozygous SNVs identified across 409 autosomal dominant disease genes with putative or established gain-of-function or dominant-negative mechanisms. Genes were sourced from Gene2Phenotype, LoGoFunc's Human Gene Mutation Database, ClinVar, and a literature review. Common single nucleotide variants (SNVs; minor allele frequency >1%) spanning the full genomic extent of these genes were extracted from the 1000 Genomes Project phase 3 dataset, representing 2,594 unrelated individuals across five super populations: African (AFR), Admixed American (AMR), East Asian (EAS), European (EUR), and South Asian (SAS). SNVs located in regions likely unsuitable for ASO-based approaches, including repetitive sequences, Alu elements, and tandem repeats, were excluded prior to analysis.
A greedy algorithm was applied to iteratively prioritize SNVs that maximize cumulative heterozygous population coverage per gene, until either ≥95% coverage or saturation was reached. For each prioritized SNV, candidate allele-specific 15-mer and 20-mer ASO sequences targeting both the reference and alternate alleles were designed in silico and filtered for appropriate sequence composition, target accessibility, and off-target effects. SNVs can be searched by gene within this tool, and corresponding ASO sequences (if available) can be downloaded for experimental evaluation.
