Skip to contents

TailQuant and associated files on disk

Load with load_tq(directory) to obtain a TailQuant object.

@sites

  • site = A unique identifier for the site.
  • location = Location in convenient string form.
  • chr = Chromosome name (or contig, patch, etc, depening on assembly).
  • pos = 1-based position of the final templated base in transcripts using this site.
  • strand = 1 or -1. Strand of transcripts using this site.
  • relation = Where the site is within the gene/transcript.
  • gene_id = Unique identifier for gene (eg ENSEMBL ID).
  • name = Name/symbol for the gene.
  • biotype = biotype of the gene.
  • product = Text description of the function of the gene, protein it produces, etc.
  • tail_counts = List of data frames, giving total read and UMI counts vs tail length.
  • various statistics

@samples

  • sample = A unique ID for the sample.
  • reads = data frames
    • chr, pos, strand, length, tail_start, tail, umi
  • sited_reads = data frames
    • site, close_to_site, chr, pos, strand, length, tail_start, tail, umi
  • tail_counts = Data frames. Summarized data about tail lengths. Events are UMIs (n_event) or reads (n_read_event).
    • site, tail, n_event, n_died, n_read_event, n_read_died
  • counts = Data frames. Summarized data about UMIs, including reads without tails.
    • site, n, n_read
  • color = (optional) If present, used in Shiny app for plotting.

/cache

Cached aggregations/computations from above.