The WALT mapper is deprecated and no longer maintained. We created a novel WGBS mapper that adds various improvements to WALT called abismal. Please use this program instead
WALT (Wildcard ALignment Tool) is a read mapping program for bisulfite sequencing in DNA methylation studies.
Currently WALT requires a C++ compiler that supports the C++11 standard. The default compiler assumed is g++ (from GCC, available on your Linux or OS X machine). The g++ compiler has supported the C++11 standard since roughly 2012 (GCC 4.7) so this should not cause any problems. WALT also requires OMP libraries and headers to be available, which rarely causes problems.
If you have trouble with the "make" part of the installation procedure described below, please contact us.
(1) Download the source code from Github
git clone git@github.com:smithlabcode/walt.git
(2) Build and Install
cd walt
make all
make install
makedb -c <genome folder or file> -o <index file>
single-end reads
walt -i <index file> -r <read files> -o <output file> [options]
paired-end reads
walt -i <index file> -1 <mate_1 read files> -2 <mate_2 read files> -o <output file> [options]
| Option | Long Tag | Type | Default | Brief Description |
|---|---|---|---|---|
| -i | -index | String | NULL | index file created by makedb command ( .dbindex) |
| -r | -reads | String | NULL | list of single-end read files (.fastq or .fq) |
| -1 | -reads1 | String | NULL | list of paired-end read _1 files (.fastq or .fq) |
| -2 | -reads2 | String | NULL | list of paired-end read _2 files (.fastq or .fq) |
| -o | -output | String | NULL | output file name (.sam or .mr) |
| -m | -mismatch | Integer | 6 | maximum allowed mismatches |
| -N | -number | Integer | 1000000 | number of reads to map in one loop |
| -a | -ambiguous | Boolean | false | randomly output one mapped position for ambiguous reads |
| -u | -unmapped | Boolean | false | output unmapped reads |
| -C | -clip | String | empty | clip the specified adaptor |
| -A | -ag-wild | Boolean | false | map using A/G bisulfite wildcards |
| -P | -pbat | Boolean | false | map post-bisulfite adaptor tagging reads |
| -b | -bucket | Integer | 5000 | maximum candidates for a seed |
| -k | -topk | Integer | 50 | maximum allowed mappings for a read in paired-end mapping |
| -L | -fraglen | Integer | 1000 | max fragment length in paired-end mapping |
| -t | -thread | Integer | 1 | number of threads for mapping |
To see the list of options, use "-?" or "-help".
(1) Indexing Genome
For example, to make an index for UCSC hg19
makedb -c hg19/ -o hg19.dbindex
or to make an index for chromosome 2
makedb -c chr2.fa -o chr2.dbindex
The suffix of the index file should be '.dbindex'.
(2) Bisulfite Mapping
For example, to mapping reads to human genome hg19
walt -i hg19.dbindex -r read_1.fq -o reads_1_mapping.sam
If mapping the reads from the *_2 reads file, the -A option should be set. This means that all Gs in the reads and genome are converted to As. If -A option is not set, all Cs in the reads and genome are converted to Ts.
walt -i hg19.dbindex -r read_2.fq -A -o reads_2_mapping.sam
If mapping post-bisulfite adaptor tagging reads (PBAT), the -P option should be set.
walt -i hg19.dbindex -r read_2.fq -P -o reads_2_mapping.sam
walt -i hg19.dbindex -1 read_1.fq -2 read_2.fq -P -o paired_mapping_PBAT.sam
Additionally, WALT supports comma-separated list of read files. WALT
produces one mapping output file for each read file. For single-end
mapping, the output file names will be appended "_s1", "_s2", and so
on. Notice: except the first file path, all other file paths cannot
use '~'. For example, "-r /read_file1.fq,/read_file2.fq" is not
allowed. It should be "-r ~/read_file1.fq,/home/read_file2.fq", since
linux system doesn't know it is a path except the first one.
walt -i hg19.dbindex -r read_file1.fq,read_file2.fq,read_file3.fq -o reads_1_mapping.sam
For paired-end reads, -1 and -2 options are used for the mate read files.
walt -i hg19.dbindex -1 read_1.fq -2 read_2.fq -o paired_mapping.sam
Similarly, WALT supports comma-separated list of read files for paired-end mapping. WALT produces one mapping output file for each pair of read files. The output file names will be appended "_p1", "_p2", and so on. One other thing to note is the mate 1 and mate 2 paired files should be in the same order.
walt -i hg19.dbindex -1 read_file1_1.fq,read_file2_1.fq,read_file3_1.fq \
-2 read_file1_2.fq,read_file2_2.fq,read_file3_2.fq \
-o paired_mapping.sam
The default number of maximum allowed mismatches is 6. The maximum allowed mismatches can be set using -m option.
walt -i hg19.dbindex -r read_1.fq -m 4 -o reads_1_mapping.sam
The option -N sets the number of reads to mapping in each loop. If N is larger, the program takes large memory, especially for paired-end mapping. For Human Genome, if N is 1000000, both single-end and paired-end mapping take about 16 Gb memory. If N is 5000000, single-end mapping takes about 18 Gb memory, while paired-end mapping takes about 28 Gb memory. If N is set to be larger than 10000000, the program will set N to be 10000000 since when N is too large the program will take large memory but it will not be significantly faster. The estimate memory for single-end mapping is 15 + N * (2 * rl + rnl + 16) / (1024^3) Gb, and for paired-end mapping is 15 + N * (4 * rl + 2 * rnl + 24 * k + 16) / (1024^3) Gb. The index size of human genome is about 15 Gb. N is the number of reads to map in one loop. rl is the length of reads (WALT supports mix of different lengths reads, so here rl is estimation of the average length). rnl is the length of read names. k is maximum allowed mappings for a read in paired-end mapping.
walt -i hg19.dbindex -r read_1.fq -N 1000000 -o reads_1_mapping.sam
To output the ambiguous mapped reads or unmapped reads, -u and -a options should be set. If the output format is MR, the ambigous mapped reads and unmapped reads will output to a separated file, respectively. For paired-end mapping, there will be a unmapped file and an ambiguous file for each of the mate 1 and mate 2 reads file. If the output format is SAM, uniquely mapped, ambiguous mapped, unmapped reads are output to the same file with different FLAGs.
walt -i hg19.dbindex -r read_1.fq -u -a -o reads_1_mapping.sam
To trim 3' end adaptor sequence, -C option should be set. For paired-end read mapping, using T_adaptor[:A_adaptor] format to set the adaptor. If only one adaptor seqeunce is given, the same adaptor sequence will be used for both T-rech and A-rich reads.
walt -i hg19.dbindex -r read_1.fq -C AGATCGGAAGAGC -o reads_1_mapping.sam
walt -i hg19.dbindex -r read_2.fq -A -C AGATCGGAAGAGC -o reads_2_mapping.sam
walt -i hg19.dbindex -1 read_1.fq -2 read_2.fq -C AGATCGGAAGAGC -o paired_mapping.sam
walt -i hg19.dbindex -1 read_1.fq -2 read_2.fq -C AGATCGGAAGAGC:AGATCGG -o paired_mapping.sam
By default, WALT produces SAM format files. If you plan to get MR file, please make sure the suffix of the output file is ".mr".
walt -i hg19.dbindex -1 read_1.fq -2 read_2.fq -o paired_mapping.mr
WALT produces a separate mapping statistics file (.mapstats) for each run. It includes percentages of uniquely mapped, ambiguously mapped and unmapped reads.
WALT supports Tab-delimited SAM and MR output formats. By default, WALT produces SAM format output files. To get MR format, the suffix of the output file should be ".mr".
SAM Format
- QNAME (read name)
- FLAG (bitwise FLAG)
- RNAME (chromosome name)
- POS (start position, 1-based, 0 means unmapped)
- MAPQ (255 in WALT)
- CIGAR (CIGAR string)
- RNEXT (chromosome name of the mate read)
- PNEXT (start position of the mate read)
- TLEN (observed segment length)
- SEQ (read sequence)
- QUAL (quality sequence read from fastq file)
- NM-tag (number of mismatches)
By default, WALT only outputs uniquely mapped reads. If -u or -a option is set, the ambiguous mapped or unmapped reads will be in the same SAM file with different FLAGs. The FLAG 0x100 will be set for ambiguous mapped reads, and the FLAG 0x4 will be set for unmapped reads.
The Mapped Read (MR) Format
This format retains less total information than is typically in a BAM file, but has all the information required for typical downstream DNA methylation analysis.
- RNAME (chromosome name)
- SPOS (start position, 0-based)
- EPOS (end position, 0-based)
- QNAME (read name)
- MISMATCH (number of mismatches)
- STRAND (forward or reverse strand)
- SEQ
- QUAL
If paired-end reads are mapped in proper pair, the QNAME is added "FRAG:" in the beginning of the read name, the STRAND is the strand of the first mate mapped and SEQ and QUAL is merged according to their mapping positions. The overlap segment of SEQ and QUAL is from the mate 1 or mate 2 and it is the one with less number of 'N' in the read sequence. MISMATCH is the sum of mismatches in the mate 1 and mismatches in the mate 2. If paired-end reads are not mapped in proper pair, they are treated as single-end reads. If -u and/or -a option is set, the ambiguous mapped or unmapped reads are output in separate files.
Haifeng Chen haifengc@usc.edu
Andrew Smith andrewds@usc.edu
Ting Chen tingchen@usc.edu
WALT: fast and accurate read mapping for bisulfite sequencing. Bioinformatics, 32(22), pp.3507-3509. Supplementary
Copyright (C) 2015 University of Southern California
Andrew D. Smith and Ting Chen
Authors: Haifeng Chen, Andrew D. Smith and Ting Chen
WALT 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.
WALT 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 WALT. If not, see <http://www.gnu.org/licenses/>.