SYNOPSIS
sim_sRNA-seq.pl
Simulation of plant small RNA-seq data
Copyright (C) 2014 Michael J. Axtell
AUTHOR
Michael J. Axtell, Penn State University, mja18@psu.edu
LICENSE
This program is free software: you can redistribute it and/or modify it
under the terms of the GNU General Public License as published by the
Free Software Foundation, either version 3 of the License, or (at your
option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <http://www.gnu.org/licenses/>.
DEPENDENCIES
perl 5 <www.perl.org> at /usr/bin/perl
samtools <http://samtools.sourceforge.net/> in the PATH
INSTALL
Ensure dependicies are present (see above). Add the script
sim_sRNA-seq.pl to your PATH
USAGE
sim_sRNA-seq.pl [options] -s soft-masked-genome.fa
--- OR, LESS PREFERRED ---
sim_sRNA-seq.pl [options] -d hard-masked-genome.fa -n not-masked-genome.fa
OPTIONS
Options:
-s : path to soft-masked genome. Lower-case letters are assumed
repeat-masked. If -s is specified, neither -d nor -n can be specified.
-d : path to hard-masked genome. If -d is specified, -n must also be
specified, and -s cannot be specified.
-n : path to not-masked genome. If -n is specified, -d must also be
specified, and -s cannot be specified.
-v : print version number and quit.
-h : print help message and quit.
-r : desired number of reads, in millions. Default: 5.
-e : per-read probability of a single nt sequencing error
(substitution). Default: 1E-4.
METHODS
Read numbers and types
30% of the simulated reads will come from roughly 100 MIRNA loci, 5% of
the simulated reads will come from roughly 20 tasiRNA/phased secondary
siRNA loci, and the remaining 65% from roughly 10,000 heterochromatic
siRNA loci.
The abundance of reads from each locus is distributed on a log-linear
scale (e.g., plotting the log10 of read number as a function of
abundance rank yields a straight line).
MIRNA simulation
Valid MIRNA loci have no overlap with repeat-masked regions of the
genome. The locus has a hypothetical size of 125nts, of which all must
be ATGC bases (e.g. no ambiguous nts.).
The strand of the MIRNA precursor is randomly selected, as is the arm
from which the mature miRNA and star come from. There is no actual
hairpin sequence necessarily present at MIRNA loci .. it is only the
pattern of reads that is being simulated.
The mature miRNA and mature miRNA* are defined as 'master' positions.
The left-most 'master' position in a locus is a 21-mer starting at
position 17 of the locus. The right-most 'master' position is a 21 mer
at position 85. Assignment of the arm (i.e., whether the left-most or
right-most 'master' position is the miRNA or miRNA*) is random at each
locus.
Once a locus has been found and 'master' positions defined, each read is
simulated according to the following probabilities. In the following
list, "miR" means mature miRNA, "star" means miRNA*. The numbers after
each indicate the offset at 5' and 3' ends relative to the master
positions. So, "miR0:1" means the mature miRNA sequence, starting at the
master 5' end, and ending 1 nt after the master 3' end.
60% miR0:0, 20% star0:0,, 4% miR0:-1, 1% miR0:-2, 4% miR1:1, 1% miR1:0,
2% miR-1:-1, 0.5% miR-1:-2, 2% miR-2:-2, 0.5% miR-2:-3, 1% star0:-1,
0.5% star0:-2, 1% star1:1, 0.5% star1:0, 2% star-1:-1.
Sampling of simulated MIRNA-derived reads continues until the required
number of reads for a particular locus is recovered.
tasiRNA/phased siRNA simulation
TAS loci have no overlap with repeat-masked areas of the genome. Each
locus has a nominal size of 140nts. All bases must be ATGC
(non-ambiguous).
Each locus is simulated to be diced in 6 21 nt phases. At each phasing
position, 21mers are the dominant size, with 20mers and 22mers being
less frequent .. the 20 and 22nt variants vary in their 3' positions
relative to the 'master' 21nt RNAs.
Once a locus has been identified, and all possible 20, 21, and 22mers
charted, each read is simulated according to the following
probabilities:
Strand of origin is 50% top, 50% bottom.
Phase position is equal chance for all (e.g. 1/6 chance for any
particular phase location).
80% of the time, the 21mer is returned, 10% of the time the 20mer, and
10% the 22 mer.
Heterochromatic siRNA simulation
Heterochromatic siRNA loci can come from anywhere in the genome
(repeat-masked or un-masked), and their nominal locus size is 100nts.
All nts in the 100 nt locus must be ATGC (no ambiguous positions).
All possible start and stop positions for 24, 23, 22, and 21 nt RNAs are
eligible from these loci. Each simulated read is identified with the
following probabilities:
50% chance for top of bottom strand origin
5' position within the locus is random
90% chance of a 24 mer, 5% chance of a 23 mer, 3% chance of a 22 mer,
and 2% chance of a 21 mer.
Simulation of sequencing errors
Regardless of small RNA type, once a given read has been simulated,
there is a chance of introucing a single-nucleotide substitution
relative to the reference genome at a randomly selected position in the
read. The chance of introducing this error is given by option -e
(Default is 1 in 10,000).
No overlapping loci
None of the simulated loci are allowed to have any overlap with each
other.
OUTPUT
Files
Two files are created in the working directory.
simulated_sRNA-seq_reads_overview.txt : A tab-delimited text file giving
the coordinates and names for each simulated locus.
simulated_sRNA-seq_reads.fasta : A FASTA file of all of the simulated
reads.
Naming conventions
The FASTA header of each simulated read encode the basic information
about the read. Several different fields are separated by "_"
characters. For instance the following read header
">MIRNA_2_2_chr7:123063894-123063914_+_0" means ..
MIRNA: This came from a simulated MIRNA locus
2: The 2nd simulated locus of this type
2: Read number 2 from this locus
chr7:123063894-123063914: The true origin of this read.
+: The genomic strand of this read.
0: The number of sequencing errors simulated into the read.