Installation¶

Test and install pipeline software¶

• Type “which gcc g++ java make gs convert pdflatex R cython” to check the installation.

After solving the dependent prerequisites, install chilin as followings,

git clone http://github.com/cfce/chilin/
cd chilin
python setup.py install -f

Then, check your installation:

source chilin_env/bin/activate
# check ChiLin dependent software and data
python setup.py -l

If any software can not be installed, look into their official documentation. Most of time, see dependentsoft to check whether all prerequisites are installed or not, usually it’s the problem of numpy, cython or gcc compiler problem, or R package seqLogo problem. Take a look at all software in the software directory to see what’s going on, and try apt-get, yum, port and pypi to fix the issue.

Lastly, user may need to check the installation of mdseqpos dependency of R seqLogo package, open R console and install dependent R packages:

R -e "source('http://bioconductor.org/biocLite.R');biocLite('seqLogo');library(seqLogo)"

Remember to source your python virtual environment “source ${ChiLin_ROOT}/chilin_env/bin/activate” everytime or put them into your ${HOME}/.bashrc or ${HOME}/.bash_profile.

Download dependent data for hg38, hg19, mm9, or mm10¶

under the ChiLin source code root directory,

# download from our cistrome server
mkdir -p db

# change directory to db
cd db

# download the one you need, this would be over 10 GB, make sure your internet access is over 100k/s, or it's too slow..
# human
wget -c http://cistrome.org/chilin/_downloads/hg19.tgz
wget -c http://cistrome.org/chilin/_downloads/hg19.tgz.md5 ## check md5
#wget -c http://cistrome.org/chilin/_downloads/hg38.tgz
#wget -c http://cistrome.org/chilin/_downloads/hg38.tgz.md5

# mouse
#wget -c http://cistrome.org/chilin/_downloads/mm9.tgz
#wget -c http://cistrome.org/chilin/_downloads/mm9.tgz.md5
#wget -c http://cistrome.org/chilin/_downloads/mm10.tgz
#wget -c http://cistrome.org/chilin/_downloads/mm10.tgz.md5

# check the md5sum for completeness of hg19
md5sum -c hg19.tgz
tar xvfz hg19.tgz

# download mycoplasma that you are afraid of contaminating your samples
wget -c http://cistrome.org/chilin/_downloads/mycoplasma.tgz
wget -c http://cistrome.org/chilin/_downloads/mycoplasma.tgz.md5
md5sum -c mycoplasma.tgz.md5
tar xvfz mycoplasma.tgz

# change back
cd ..

# check your data and software installation, if download is ok
python setup.py -l

If you

• want to know more about dependent data
• want to prepare new version of reference data by yourself
• you have species assembly not in our list

see details about the dependent data.

add species support in chilin¶

After these preparation of software and reference data, if you are using our prepared hg38, hg19, mm10, mm9 dependent data, you can skip this part because setup.py already sets chilin.conf.filled for you. If you have your own reference data, open your favorite text editor, appending section in chilin.conf.filled file add in species support like this.

fill in the section with your own data absolute path, then append filled following section

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32

[hg19]
genome_index =
# fasta file separated by chromosome, such as chr1.fa
genome_dir =
chrom_len =
dhs =
## blacklist region
velcro =
conservation =
geneTable =

[mm9]
genome_index =
# fasta file separated by chromosome, such as chr1.fa
genome_dir =
chrom_len =
dhs =
## blacklist region
velcro =
conservation =
geneTable =

[hg38]
genome_index =
# fasta file separated by chromosome, such as chr1.fa
genome_dir =
chrom_len =
dhs =
## blacklist region
velcro =
conservation =
geneTable =

after conf, see details about this in [species].

#Define the site-wide defaults for your system using ABSOLUTE PATHS
#When defining reference species (i.e. your species of interest)
#please refer to "Generating Species References" in README.md on how
#to generate the files
#------------------------------------------------------------------------------
# Tools
#------------------------------------------------------------------------------
#ChiLin is dependent on several tools, please specify the absolute path to
#these tools--ALL FIELDS ARE REQUIRED
#put bedClip, bedGraphToBigWig, bowtie, star, bwa, fastqc, bedtools, macs2, samtools, seqtk, wigCorrelate
#in executable PATH
#other system tool includes convert, pdflatex, R, python2.7
[tool]
mdseqpos =
macs2 =

#------------------------------------------------------------------------------
# Tool parameters
#------------------------------------------------------------------------------
#These are optional parameters for some tools defined above
#NOTE: not all tool parameters can be inputted in this conf--e.g. for bwa,
#the only thing you can affect here is the number of threads used.
[macs2]
#refer to the macs2 help message to find out what these mean, species for effective genome size
extsize = 146
# effecitive genome sizes, support hs, mm, other species, please refer to chromInfo
species = 
type = both
fdr = 0.01
keep_dup = 1

[reg]
## regulatory potential score prediction top peaks
peaks = 10000
dist = 100000

[conservation]
## for tf/dnase we suggest 400bp width around summit, for histone 4000
type = tf
peaks = 5000
width = 400

[seqpos]
peaks = 5000
mdscan_width = 200
mdscan_top_peaks = 200
seqpos_mdscan_top_peaks_refine = 500
width = 600
pvalue_cutoff = 0.001
db = cistrome.xml

#------------------------------------------------------------------------------
# Contamination
#------------------------------------------------------------------------------
#OPTIONAL- our contamination module can screen for any species defined below
#specify the species name and the path to the bwa index as follows: e.g.
#ECOLI = /some/path/ecoli
[contamination]
mycoplasma = mycoplasma
# ecoli =
# yeat = 

Test installation¶

Test installation with demo data,

# non cluster server
cd demo
bash foxa1

# if you are using slurm sytem
cd demo
# submit cluster script `foxa1`
sbatch foxa1

Check demo data results,

du -h local/local.pdf ## quality report
# mac
open local/local.pdf
# linux
nautilus local/

For more options, see Manual.