Skip to content

Output

k4neo generates different output files depending on the subcommand used. This page describes the generated files and their contents.

k4neo-annotator

k4neo-annotator \
  --database tests/resources/index_metadata.db \
  --index tests/resources/index_manifest.yaml \
  --queries tests/resources/queries/test_junction_k4neo_input.tsv \
  --working-dir . \
  --output test_jx

Note: --output specifies an output prefix, not a filename.

This generates four output files in your current working directory. By default, the main output file aggregates all hits by tissue, disease, and developmental_stage per study (study-specific), while the sample rate outputs aggregate by tissue, disease, and developmental_stage combinations (tissue-specific).

  • test_jx_annotated_raptor.tsv.gz: Lists all input sequences with hits per tissue and developmental stage for each indexed study:
    • cts_id: Unique sequence identifier (from input).
    • count: Number of samples expressing the sequence with at least x% k-mer ratio (default: 70%).
    • total: Total number of samples in the study.
    • disease: Disease status of the original RNA-seq samples.
    • developmental_stage: Developmental stage of the original RNA-seq samples.
    • tissue: Tissue type of original RNA-seq samples.
    • study_id: Unique study identifier.

Example output:

cts_id count total disease developmental_stage tissue study_id
46971 2 3 healthy adult brain E-MTAB-2836

Sequence 46971 was found in 2 out of 3 healthy adult brain samples from study E-MTAB-2836.

  • test_jx_healthy_sample_rate_raptor.tsv.gz: Contains the healthy tissue sample rate for each sequence:

    • cts_id: Unique sequence identifier (from input).
    • developmental_stage: Developmental stage description of original RNA-seq samples.
    • tissue: Tissue type of original RNA-seq samples.
    • sample_rate: Fraction of samples with ≥ x% k-mer coverage along the sequence (e.g., 0.7 = 70% of aggregated samples show expression).

  • test_jx_tumor_sample_rate_raptor.tsv.gz: Contains the tumor entity sample rate for each sequence:

    • cts_id: Unique sequence identifier (from input).
    • disease: Disease status of the sample (TCGA index values: disease, metastatic, primary blood tumor, primary solid tumor).
    • tissue: Cancer type of samples.
    • sample_rate: Fraction of samples with ≥ x% k-mer coverage along the sequence (e.g., 0.7 = 70% of aggregated tumor samples show expression).

  • test_jx_cts_to_query_cts.tsv: Contains a mapping between user-provided sequence IDs and k4neo internal sequence IDs:

    • cts_id: Sequence ID provided in input.
    • query_cts_id: Sequence ID used internally by k4neo.

k4neo-uniq

k4neo-uniq \
  --fasta query.fa \
  --reference_indices tests/resources/index/ref_meta.json \
  --output uniq_annot.tsv

This generates a file called uniq_annot.tsv with the following annotation columns:

  • cts_id: Query sequence identifier.
  • cts_unique_rate: Rate of k-mers specific to the query sequence. For novel variants, this should ideally exceed the k-mer fraction used for search.
  • cts_ref_rate: Rate of k-mers occurring at least once in the reference genome or transcriptome.
  • cts_ref_single_gene_locus_rate: Rate of k-mers occurring only once in a genic region of the reference.
  • cts_ref_multi_gene_locus_rate: Rate of k-mers occurring in multiple genic regions of the reference (e.g., through repeats or paralogous regions).
  • cts_ref_single_transcript_rate: Rate of k-mers occurring at only one genic region and specific to a single isoform of that locus.
  • cts_ref_multi_transcript_rate: Rate of k-mers occurring at only one genic region but present in multiple isoforms of that locus.

k4neo-quant

k4neo-quant \
  --index /path/to/quant_index.yaml \
  --fasta query.fa \
  --output quant_annotation.tsv \
  --cpu 2 \
  --normalize

This generates a file called quant_annotation.tsv with the following annotation columns:

  • cts_id: Query sequence identifier.
  • sample: Identifier of the indexed sample.
  • median_kmer_count: Median count of query k-mers.
  • mean_kmer_count: Mean count of query k-mers.
  • max_kmer_count: Maximum k-mer count of query.
  • min_kmer_count: Minimum k-mer count of query.
  • rate_non_zero_kmers: Rate of k-mers with counts in the index.
  • rate_zero_kmers: Rate of k-mers with zero count in the index.
  • variance: Variance of k-mer counts (indicates coverage uniformity).
  • cv: Coefficient of variation.