detailed_manual.md

Terminology: breakpoint types and strandedness

• Breakpoint type:

  • S2G: telomere-containing softclipped-reads align to a location in the genome
  • G2S: non-telomere-containing softclipped-reads align to a location in the genome that contains telomeres

  These two types both describe elimination breakpoints at which telomeres have been added to the retained fragments. In the case of S2G breakpoints, the assembled genome is the unbroken genome, and breakpoint-supporting reads come from cells with a broken genome. In the case of G2S breakpoints, the assembled genome is the reduced genome (with telomeres), and breakpoint-supporting reads come from cells with an unbroken genome

• Breakpoint strand:

  • '+', or also called '3prime', if the softclips on reads occur 3' of the assembled genome
  • '-' or '5prime', if they occur 5' of the assembled genome

  For both S2G and G2S breakpoints, '+' suggests the eliminated genome occurs 3' of the identified breakpoint, and vice-versa.

Output files in detail

• breakpoint_locations.bed: a BED-formatted file containing the location of identified elimination breakpoints.

  • The location (columns 1,2 and 3) is provided as an interval of size one, and is defined as the first eliminated base past the putative breakpoint. If multiple putative breakpoints have been clustered, the single location with the maximal read support is used.
  • The name column (column 4) records the breakpoint type, as defined above. It also stores a larger window, in the format breakpoint_window: <start>-<end>. This contains all putative breakpoint positions with >=--min_supporting_reads read support, and located within --clustering_threshold of each other.
  • The score column (column 5) stores the number of sequencing reads that support the breakpoint.
  • The strand column (column 6) specifies the likely direction of eliminated DNA, as defined above.

• breakpoint_sequences.fasta: a FASTA-formatted file containing the sequences of identified elimination breakpoints. The header field contains:

  • A sequence ID as <breakpoint_type>_<breakpoint_direction>_<chrom>

    breakpoint_type: 'S2G' or 'G2S', as defined above

    breakpoint_type: '5prime' or '3prime', as defined above

    chrom: the contig/scaffold/chromosome name

  • Some additional information is provided, e.g. the position of the breakpoint and the number of reads supporting the breakpoint ('num_telo_containing_softclips')

• breakpoint_foci_<breakpoint_type>.tsv: a tab-separated-value file containing the location of all putative breakpoints, the read support for each breakpoint (in both forward and reverse orientation), and the total read depth at the putative breakpoint, plus in a window around each breakpoint. This file enables assessing how sharp a breakpoint is, and accessing all the individual breakpoints that may have been clustered in breakpoint_locations.bed.

Applications

Assembling past somatic telomeres

The 'G2S' breakpoints in the output BED of delfies indicate locations at which the genome assembly stops at telomeres of the reduced genome (in species that undergo Programmed DNA Elimination, this is the somatic genome).

I provide a script in scripts/deplete_G2S_reads.sh for removing from your input BAM the reads that support these somatic telomeres. Re-assembling the 'soma-depleted' reads only allows extending past them and assembling a more complete genome.

Here is an example of reads aligned to a G2S breakpoint, before and after running the G2S read depletion script:

The genome sequence at the bottom has telomere arrays assembled 5' of an identified G2S breakpoint by delfies. The alignments in the top panel contain a mixture of 'reduced' (telomere-containing, i.e. matching the assembled genome) and 'complete' reads (softclips where the telomeres are in the genome). The alignments in the bottom panel contain 'complete' reads only. These can be assembled to 'force' the genome assembler not to stop at the reduced genome.