An alignment pipeline.
This is a fork of BSmooth by Kasper D. Hansen, Ben Langmead, and Rafael A. Irizarry. The fork fixes some problems which prevented the merman aligner from compiling.
BSmooth is by Kasper D. Hansen, Ben Langmead, Rafael A. Irizarry
Statistics pipeline is by Kasper D. Hansen Alignment pipeline is by Ben Langmead
The BSmooth alignment pipeline is licensed under the GPLv3 license. See `LICENSE' file for details.
To install the BSmooth alignment pipeline, permanently set the BSMOOTH_HOME environment variable to the main BSmooth directory. I.e., the directory with 'bin', 'example' and 'lib' subdirectories inside.
The merman sources are included with BSmooth. To use merman with BSmooth, you must first build the merman binary. This requires standard GNU build tools such as GNU make and g++.
To build merman, cd to the $BSMOOTH_HOME/merman directory and run 'make'. Finally, add the $BSMOOTH_HOME/merman directory to your PATH.
Follow the standard instructions for installing those tools; see their respective manuals for details. Once the tool is installed, make sure that the directory containing the key binaries ('bowtie2', 'bowtie2-build') is included in your PATH.
The following 4 steps (or possibly 3 steps, depending on aligner) are needed to get from raw input reads to CpG-level measurement summaries. These summaries constitute the input to the R pipeline for smoothing and calculating DMRs.
-
Build bisulfite genome index (not necessary if using Merman) bswc_bowtie2_index.pl
-
Align bs_merman_align.pl bswc_bowtie2_align.pl
-
Sort evidence directory bsev_sort.pl
-
Tabulate sorted evidence directory bsev_tabulate.pl
For an example, see example/Makefile, which uses example/sim.pl to create a small, simulated problem, then goes through the steps listed above.