Bulked segregant analysis, as implemented in MutMap (Abe et al., 2012), is a powerful and efficient method to identify agronomically important loci in crop plants. MutMap requires whole-genome resequencing of a single individual from the original cultivar and the pooled sequences of F2 progeny from a cross between the original cultivar and mutant. MutMap uses the sequence of the original cultivar to polarize the site frequencies of neighbouring markers and identifies loci with an unexpected site frequency, simulating the genotype of F2 progeny. The updated pipeline is approximately 5-8 times faster than the previous pipeline, is easier for novice users to use, and can be easily installed through bioconda with all dependencies.
Citation
Yu Sugihara, Lester Young, Hiroki Yaegashi, Satoshi Natsume, Daniel J. Shea, Hiroki Takagi, Helen Booker, Hideki Innan, Ryohei Terauchi, Akira Abe (2022). High performance pipeline for MutMap and QTL-seq. PeerJ, 10:e13170.
MutMap also checks required external commands at startup and reports missing commands before running the pipeline.
Usage
If your reference genome contains more than 50 contigs, only the 50 longest contigs will be plotted.
mutmap -h
usage: mutmap -r <FASTA> -c <BAM|FASTQ> -b <BAM|FASTQ>
-n <INT> -o <OUT_DIR> [-T] [-e <DATABASE>]
MutMap version 2.4.0
options:
-h, --help show this help message and exit
-r , --ref Reference FASTA file.
-c , --cultivar FASTQ or BAM file of cultivar. If specifying
FASTQ, separate paired-end files with a comma,
e.g., -c fastq1,fastq2. This option can be
used multiple times.
-b , --bulk FASTQ or BAM file of mutant bulk. If specifying
FASTQ, separate paired-end files with a comma,
e.g., -b fastq1,fastq2. This option can be
used multiple times.
-n , --N-bulk Number of individuals in the mutant bulk.
-o , --out Output directory. The specified directory must not
already exist.
-t , --threads Number of threads. If a value less than 1 is specified,
MutMap will use the maximum available threads. [2]
-w , --window Window size in kilobases (kb). [2000]
-s , --step Step size in kilobases (kb). [100]
-D , --max-depth Maximum depth of variants to be used. This cutoff
applies to both the cultivar and the bulk. [250]
-d , --min-depth Minimum depth of variants to be used. This cutoff
applies to both the cultivar and the bulk. [8]
-N , --N-rep Number of replicates for simulations to generate
null distribution. [5000]
-T, --trim Trim FASTQ files using Trimmomatic.
-a , --adapter FASTA file containing adapter sequences. This option
is used when "-T" is specified for trimming.
--trim-params Parameters for Trimmomatic. Input parameters
must be comma-separated in the following order:
Phred score, ILLUMINACLIP, LEADING, TRAILING,
SLIDINGWINDOW, MINLEN. To remove Illumina adapters,
specify the adapter FASTA file with "--adapter".
If not specified, adapter trimming will be skipped.
[33,<ADAPTER_FASTA>:2:30:10,20,20,4:15,75]
-e , --snpEff Predict causal variants using SnpEff. Check
available databases in SnpEff.
--line-colors Colors for threshold lines in plots. Specify a
comma-separated list in the order of SNP-index,
p95, and p99. ["#D55D00,#009E72,#FDB003"]
--dot-color Color of the dots in plots. ["#B3B8DD"]
--mem Maximum memory per thread when sorting BAM files;
suffixes K/M/G are recognized. [1G]
-q , --min-MQ Minimum mapping quality for mpileup. [40]
-Q , --min-BQ Minimum base quality for mpileup. [18]
-C , --adjust-MQ Adjust the mapping quality for mpileup. The default
setting is optimized for BWA. [50]
-v, --version show program's version number and exit
MutMap can be run from FASTQ (without or with trimming) and BAM. If you want to run MutMap from VCF, please use MutPlot (example 5). Once you run MutMap, MutMap automatically completes the subprocesses.
-c : FASTQs of cultivar. Please input paired-end reads separated by a comma. FASTQ files can be gzipped.
-b : FASTQs of mutant bulk. Please input paired-end reads separated by a comma. FASTQ files can be gzipped.
-n : number of individuals in mutant bulk.
-o : name of output directory. The specified directory should not already exist.
-T : trim your reads by trimmomatic.
-a : FASTA of adapter sequences for trimmomatic.
If you are using TrueSeq3, you can find the adapter sequences in the Github page of Trimmomatic. This thread is also helpful to preprare the adapter file.
MutMap automatically merges multiple FASTQ and BAM files. Of course, you can merge FASTQ or BAM files using cat or samtools merge before inputting them into MutMap. If you specify -c multiple times, please make sure that those files include only one individual. On the other hand, -b can include more than one individual because they are bulked samples. MutMap automatically classifies FASTQ and BAM files based on whether comma exists or not.
Example 5 : run MutPlot from VCF
usage: mutplot -v <VCF> -o <OUT_DIR> -n <INT> [-w <INT>] [-s <INT>]
[-D <INT>] [-d <INT>] [-N <INT>] [-m <FLOAT>]
[-S <INT>] [-e <DATABASE>] [--igv] [--indel]
MutPlot version 2.4.0
options:
-h, --help show this help message and exit
-v , --vcf VCF file which contains cultivar and mutant bulk.
in this order. This VCF file must have AD field.
-o , --out Output directory. The specified directory can already
exist.
-n , --N-bulk Number of individuals in the mutant bulk.
-w , --window Window size in kilobases (kb). [2000]
-s , --step Step size in kilobases (kb). [100]
-D , --max-depth Maximum depth of variants to be used. This cutoff
applies to both the cultivar and the bulk. [250]
-d , --min-depth Minimum depth of variants to be used. This cutoff
applies to both the cultivar and the bulk. [8]
-N , --N-rep Number of replicates for simulations to generate
null distribution. [5000]
-m , --min-SNPindex Cutoff of minimum SNP-index for clear results. [0.3]
-S , --strand-bias Filter out spurious homozygous genotypes in the cultivar
based on strand bias. If ADF (or ADR) is higher than
this cutoff when ADR (or ADF) is 0, that SNP will be
filtered out. If you want to disable this filtering,
set this cutoff to 0. [7]
-e , --snpEff Predict causal variants using SnpEff. Check
available databases in SnpEff.
--igv Output IGV format file to check results on IGV.
--indel Plot SNP-index with INDEL.
--line-colors Colors for threshold lines in plots. Specify a
comma-separated list in the order of SNP-index,
p95, and p99. ["#D55D00,#009E72,#FDB003"]
--dot-color Color of the dots in plots. ["#B3B8DD"]
--fig-width Width allocated in chromosome figure. [7.5]
--fig-height Height allocated in chromosome figure. [4.0]
--white-space White space between figures. (This option
only affects vertical direction.) [0.6]
-f , --format Specify the format of an output image.
eps/jpeg/jpg/pdf/pgf/png/rgba/svg/svgz/tif/tiff
--version show program's version number and exit
MutPlot is included in MutMap. MutMap runs MutPlot after making the VCF. Then, MutPlot will work with default parameters. If you want to change some parameters, you can use VCF inside of (OUT_DIR/30_vcf/mutmap.vcf.gz) to retry plotting process like below.
If you run MutMap with trimming, you will get the directory of 00_fastq which includes FASTQs after trimming.
You can check the results in 40_mutmap.
snp_index.tsv : columns in this order.
CHROM : chromosome name
POSI : position in chromosome
VARIANT : SNP or INDEL
DEPTH : depth of bulk
p99 : 99% confidence interval of simulated SNP-index
p95 : 95% confidence interval of simulated SNP-index
SNP-index : real SNP-index
sliding_window.tsv : columns in this order.
CHROM : chromosome name
POSI : central position of window
MEAN p99 : mean of p99
MEAN p95 : mean of p95
MEAN SNP-index : mean SNP-index
snp_index.p95.tsv and snp_index.p99.tsv contain only the SNPs that exceed the respective thresholds (95% or 99%). Similarly, sliding_window.p95.tsv and sliding_window.p99.tsv contain only the windows that exceed the respective thresholds.
mutmap_plot.png : resulting plot (like below)
BLUE dot : variant
RED line : mean SNP-index
ORANGE line : mean p99
GREEN line : mean p95
If you run MutMap with SnpEff, the following additional outputs will be generated:
mutmap.snpEff.vcf: The updated VCF file after annotation by SnpEff, located in the 40_mutmap directory.
snp_index.tsv: This file will contain a new column, impact, which includes the mutation impact information predicted by SnpEff.
When plotting the results, variants classified as MODERATE by SnpEff are marked with a + symbol, while variants classified as HIGH are marked with an x symbol in the plot.
Additional Resources
MutMap commands breakdown
For a detailed breakdown of the commands used in MutMap, including explanations of each step, parameters, and best practices for troubleshooting, please refer to the MutMap Commands Breakdown document.
Build a custom SnpEff database
If you are working with a non-model organism or your own reference genome, you may need to build a custom SnpEff database. For detailed instructions on how to build a custom SnpEff database, please refer to the Build a Custom SnpEff Database document.
Frequently Asked Questions
Choosing the reference genome
You can use a line that was not involved in the cross as the reference genome. In the version of MutMap published by Abe et al., 2012, the reference fasta was rebuilt using the wild-type cultivar’s reads to create a consensus sequence. However, starting from version 2, that step has been omitted. Instead, the VCF file is used to determine which mutations are derived from the wild-type cultivar and which are from the mutant.
When there are many contigs
The current setting has been updated to pick the top 50 contigs based on length, so only these contigs will be displayed in the plot. However, the table contains SNP-index information for all contigs, allowing you to confirm significant SNPs even for contigs not shown in the plot. You can also generate plots for these SNPs independently if needed. Since contigs smaller than the sliding window size often produce less reliable results, excluding them from the analysis should not be an issue.
How to filter for higher-confidence SNPs
If the phenotype is clear and the sequence data is clean, the default settings should already show some results. If you want to focus on higher-confidence SNPs, you can sequence both parents and retain only the SNPs that are clearly 0/0 in one parent and 1/1 in the other in the VCF file, which should produce cleaner results. However, in MutMap, keep in mind that you cannot input the mutant parent’s sequence—only the sequence of the original cultivar can be used as input. The linked page explains default MutMap commands, which involve only one parent in the workflow, but it may still be helpful as a reference when creating VCFs that include both parents.
About multiple testing correction
This function has been deprecated since v2.3.5. We highly recommend running MutMap without this function. However, if you would like to use this function, you can use it with versions of MutMap older than v2.3.5.
Troubleshooting common errors
If MutPlot reports that no variants passed filters, relax one or more filters and retry:
-d/--min-depth
-D/--max-depth
-m/--min-SNPindex
-S/--strand-bias
If you specify colors manually, --line-colors must contain exactly 3 comma-separated values in this order:
SNP-index, p95, p99
If MutMap stops before running commands, check whether required external tools are installed and in your PATH:
bwa, samtools, bcftools, tabix (plus trimmomatic when -T, and snpEff when -e)
If plotting fails because of crowded panels or clipped labels, increase --fig-width / --fig-height and rerun mutplot using OUT_DIR/30_vcf/mutmap.vcf.gz.
MutMap User Guide
version 2.4.0
Table of contents
What is MutMap?
Bulked segregant analysis, as implemented in MutMap (Abe et al., 2012), is a powerful and efficient method to identify agronomically important loci in crop plants. MutMap requires whole-genome resequencing of a single individual from the original cultivar and the pooled sequences of F2 progeny from a cross between the original cultivar and mutant. MutMap uses the sequence of the original cultivar to polarize the site frequencies of neighbouring markers and identifies loci with an unexpected site frequency, simulating the genotype of F2 progeny. The updated pipeline is approximately 5-8 times faster than the previous pipeline, is easier for novice users to use, and can be easily installed through bioconda with all dependencies.
Citation
Yu Sugihara, Lester Young, Hiroki Yaegashi, Satoshi Natsume, Daniel J. Shea, Hiroki Takagi, Helen Booker, Hideki Innan, Ryohei Terauchi, Akira Abe (2022). High performance pipeline for MutMap and QTL-seq. PeerJ, 10:e13170.
Akira Abe, Shunichi Kosugi, Kentaro Yoshida, Satoshi Natsume, Hiroki Takagi, Hiroyuki Kanzaki, Hideo Matsumura, Kakoto Yoshida, Chikako Mitsuoka, Muluneh Tamiru, Hideki Innan, Liliana Cano, Sophien Kamoun & Ryohei Terauchi (2012). Genome sequencing reveals agronomically important loci in rice using MutMap. Nature Biotechnol. 30:174-179.
Installation
Dependencies
Software
Python (>=3.5) libraries
Installation via bioconda
You can install MutMap via bioconda.
Installation via environment file
If you install from this repository,
environment.ymlis the easiest way to install MutMap with external dependencies in one step.MutMap also checks required external commands at startup and reports missing commands before running the pipeline.
Usage
If your reference genome contains more than 50 contigs, only the 50 longest contigs will be plotted.
MutMap can be run from FASTQ (without or with trimming) and BAM. If you want to run MutMap from VCF, please use MutPlot (example 5). Once you run MutMap, MutMap automatically completes the subprocesses.
Example 1 : run MutMap from FASTQ without trimming
-r: reference fasta-c: FASTQs of cultivar. Please input paired-end reads separated by a comma. FASTQ files can be gzipped.-b: FASTQs of bulk. Please input paired-end reads separated by a comma. FASTQ files can be gzipped.-n: number of individuals in mutant bulk.-o: name of output directory. The specified directory should not already exist.Example 2 : run MutMap from FASTQ with trimming
-r: reference fasta-c: FASTQs of cultivar. Please input paired-end reads separated by a comma. FASTQ files can be gzipped.-b: FASTQs of mutant bulk. Please input paired-end reads separated by a comma. FASTQ files can be gzipped.-n: number of individuals in mutant bulk.-o: name of output directory. The specified directory should not already exist.-T: trim your reads by trimmomatic.-a: FASTA of adapter sequences for trimmomatic.If you are using TrueSeq3, you can find the adapter sequences in the Github page of Trimmomatic. This thread is also helpful to preprare the adapter file.
Example 3 : run MutMap from BAM
-r: reference fasta-c: BAM of cultivar.-b: BAM of mutant bulk.-n: number of individuals in mutant bulk.-o: name of output directory. The specified directory should not already exist.Example 4 : run MutMap from multiple FASTQs and BAMs
MutMap automatically merges multiple FASTQ and BAM files. Of course, you can merge FASTQ or BAM files using
catorsamtools mergebefore inputting them into MutMap. If you specify-cmultiple times, please make sure that those files include only one individual. On the other hand,-bcan include more than one individual because they are bulked samples. MutMap automatically classifies FASTQ and BAM files based on whether comma exists or not.Example 5 : run MutPlot from VCF
MutPlot is included in MutMap. MutMap runs MutPlot after making the VCF. Then, MutPlot will work with default parameters. If you want to change some parameters, you can use VCF inside of
(OUT_DIR/30_vcf/mutmap.vcf.gz)to retry plotting process like below.Use MutPlot for a VCF which was made by yourself
In this case:
If you encounter an error, please try running MutMap from FASTQ or BAM before reporting it in the issues.
Outputs
Inside of
OUT_DIRis like below.00_fastqwhich includes FASTQs after trimming.40_mutmap.snp_index.tsv: columns in this order.sliding_window.tsv: columns in this order.snp_index.p95.tsvandsnp_index.p99.tsvcontain only the SNPs that exceed the respective thresholds (95% or 99%). Similarly,sliding_window.p95.tsvandsliding_window.p99.tsvcontain only the windows that exceed the respective thresholds.mutmap_plot.png: resulting plot (like below)40_mutmapdirectory.+symbol, while variants classified as HIGH are marked with anxsymbol in the plot.Additional Resources
MutMap commands breakdown
For a detailed breakdown of the commands used in MutMap, including explanations of each step, parameters, and best practices for troubleshooting, please refer to the MutMap Commands Breakdown document.
Build a custom SnpEff database
If you are working with a non-model organism or your own reference genome, you may need to build a custom SnpEff database. For detailed instructions on how to build a custom SnpEff database, please refer to the Build a Custom SnpEff Database document.
Frequently Asked Questions
Choosing the reference genome
You can use a line that was not involved in the cross as the reference genome. In the version of MutMap published by Abe et al., 2012, the reference fasta was rebuilt using the wild-type cultivar’s reads to create a consensus sequence. However, starting from version 2, that step has been omitted. Instead, the VCF file is used to determine which mutations are derived from the wild-type cultivar and which are from the mutant.
When there are many contigs
The current setting has been updated to pick the top 50 contigs based on length, so only these contigs will be displayed in the plot. However, the table contains SNP-index information for all contigs, allowing you to confirm significant SNPs even for contigs not shown in the plot. You can also generate plots for these SNPs independently if needed. Since contigs smaller than the sliding window size often produce less reliable results, excluding them from the analysis should not be an issue.
How to filter for higher-confidence SNPs
If the phenotype is clear and the sequence data is clean, the default settings should already show some results. If you want to focus on higher-confidence SNPs, you can sequence both parents and retain only the SNPs that are clearly 0/0 in one parent and 1/1 in the other in the VCF file, which should produce cleaner results. However, in MutMap, keep in mind that you cannot input the mutant parent’s sequence—only the sequence of the original cultivar can be used as input. The linked page explains default MutMap commands, which involve only one parent in the workflow, but it may still be helpful as a reference when creating VCFs that include both parents.
About multiple testing correction
This function has been deprecated since v2.3.5. We highly recommend running MutMap without this function. However, if you would like to use this function, you can use it with versions of MutMap older than v2.3.5.
Troubleshooting common errors
-d/--min-depth-D/--max-depth-m/--min-SNPindex-S/--strand-bias--line-colorsmust contain exactly 3 comma-separated values in this order:PATH:bwa,samtools,bcftools,tabix(plustrimmomaticwhen-T, andsnpEffwhen-e)--fig-width/--fig-heightand rerunmutplotusingOUT_DIR/30_vcf/mutmap.vcf.gz.