FindDelMH: Return the length of microhomology at a deletion

In ICAMS: In-Depth Characterization and Analysis of Mutational Signatures ('ICAMS')

Description

Return the length of microhomology at a deletion

Usage

FindDelMH(context, deleted.seq, pos, trace = 0, warn.cryptic = TRUE)

Arguments

context	The deleted sequence plus ample surrounding sequence on each side (at least as long as del.sequence).
deleted.seq	The deleted sequence in context.
pos	The position of del.sequence in context.
trace	If > 0, then generate various messages showing how the computation is carried out.
warn.cryptic	if TRUE generating a warning if there is a cryptic repeat (see the example).

Details

This function is primarily for internal use, but we export it to document the underlying logic.

Example:

GGCTAGTT aligned to GGCTAGAACTAGTT with a deletion represented as:

GGCTAGAACTAGTT
GG------CTAGTT GGCTAGTT GG[CTAGAA]CTAGTT
                           ----   ----

Presumed repair mechanism leading to this:

  ....
GGCTAGAACTAGTT
CCGATCTTGATCAA

=>

  ....
GGCTAG      TT
CC      GATCAA
        ....

=>

GGCTAGTT
CCGATCAA

Variant-caller software can represent the same deletion in several different, but completely equivalent, ways.

GGC------TAGTT GGCTAGTT GGC[TAGAAC]TAGTT
                          * ---  * ---

GGCT------AGTT GGCTAGTT GGCT[AGAACT]AGTT
                          ** --  ** --

GGCTA------GTT GGCTAGTT GGCTA[GAACTA]GTT
                          *** -  *** -

GGCTAG------TT GGCTAGTT GGCTAG[AACTAG]TT
                          ****   ****

This function finds:

1. The maximum match of undeleted sequence to the left of the deletion that is identical to the right end of the deleted sequence, and

2. The maximum match of undeleted sequence to the right of the deletion that is identical to the left end of the deleted sequence.

The microhomology sequence is the concatenation of items (1) and (2).

Warning
A deletion in a repeat can also be represented in several different ways. A deletion in a repeat is abstractly equivalent to a deletion with microhomology that spans the entire deleted sequence. For example;

GACTAGCTAGTT
GACTA----GTT GACTAGTT GACTA[GCTA]GTT
                        *** -*** -

is really a repeat

GACTAG----TT GACTAGTT GACTAG[CTAG]TT
                        **** ----

GACT----AGTT GACTAGTT GACT[AGCT]AGTT
                        ** --** --

This function only flags these "cryptic repeats" with a -1 return; it does not figure out the repeat extent.

Value

The length of the maximum microhomology of del.sequence in context.

ID classification

See https://github.com/steverozen/ICAMS/raw/master/data-raw/PCAWG7_indel_classification_2017_12_08.xlsx for additional information on ID (small insertion and deletion) mutation classification.

See the documentation for Canonicalize1Del which first handles deletions in homopolymers, then handles deletions in simple repeats with longer repeat units, (e.g. CACACACA, see FindMaxRepeatDel), and if the deletion is not in a simple repeat, looks for microhomology (see FindDelMH).

See the code for unexported function CanonicalizeID and the functions it calls for handling of insertions.

Examples

# GAGAGG[CTAGAA]CTAGTT
#        ----   ----
FindDelMH("GGAGAGGCTAGAACTAGTTAAAAA", "CTAGAA", 8, trace = 0)  # 4

# A cryptic repeat
# 
# TAAATTATTTATTAATTTATTG
# TAAATTA----TTAATTTATTG = TAAATTATTAATTTATTG
# 
# equivalent to
#
# TAAATTATTTATTAATTTATTG
# TAAAT----TATTAATTTATTG = TAAATTATTAATTTATTG 
# 
# and
#
# TAAATTATTTATTAATTTATTG
# TAAA----TTATTAATTTATTG = TAAATTATTAATTTATTG  

FindDelMH("TAAATTATTTATTAATTTATTG", "TTTA", 8, warn.cryptic = FALSE) # -1

ICAMS documentation built on April 3, 2021, 5:07 p.m.