GIGGLE is a genomics search engine that identifies and ranks the significance of shared genomic loci between query features and thousands of genome interval files.
Choose one of the following methods:
Conda/Mamba
mamba env create -f environment.yml
mamba activate giggle-dev
export HTS_INC=$CONDA_PREFIX/include
export HTS_LIB=$CONDA_PREFIX/libNix
nix developUbuntu/Debian
sudo apt-get install gcc make zlib1g-dev libssl-dev libhts-dev
export HTS_INC=/usr/include
export HTS_LIB=/usr/libgit clone https://github.com/ryanlayer/giggle.git
cd giggle
make# Create some test data
mkdir -p example/beds && cd example
# Download and prepare a small dataset
curl -s "http://hgdownload.soe.ucsc.edu/goldenPath/hg19/database/microsat.txt.gz" \
| gunzip -c | cut -f 2,3,4 | sort -k1,1 -k2,2n | bgzip > beds/microsat.bed.gz
curl -s "http://hgdownload.soe.ucsc.edu/goldenPath/hg19/database/simpleRepeat.txt.gz" \
| gunzip -c | cut -f 2,3,4 | sort -k1,1 -k2,2n | bgzip > beds/simpleRepeat.bed.gz
# Index the files
giggle index -i "beds/*.bed.gz" -o my_index -s -f
# Search a region
giggle search -i my_index -r chr1:1000000-2000000
# Search with full record output
giggle search -i my_index -r chr1:1000000-2000000 -vGIGGLE has two main commands: index and search.
giggle index -i <input files> -o <output dir> -f
-i Files to index (e.g. "data/*.gz")
-o Index output directory
-s Files are sorted (faster indexing)
-f Force reindex if output directory exists
-m Metadata config file (see experiments/metadata_index_query_filter)
Note: Input files must be bgzipped BED or VCF files.
giggle search -i <index directory> [options]
-i GIGGLE index directory
-r Region(s) to search (e.g. "chr1:1000-2000" or CSV "chr1:1-100,chr2:1-100")
-q Query file (bgzipped BED or VCF)
-c Show counts by indexed file
-s Show significance statistics (requires -q)
-v Show full matching records
-o Group results by query record (use with -v)
-f Filter to files matching pattern (regex CSV)
-g Genome size for significance testing (default: 3095677412)
-l List files in the index
-m Load metadata index
-u Apply query filter
Count overlaps per file:
giggle search -i my_index -r chr1:1000000-2000000
# Output:
# #microsat.bed.gz size:41572 overlaps:5
# #simpleRepeat.bed.gz size:962714 overlaps:23Get full records:
giggle search -i my_index -r chr1:1000000-2000000 -vFilter to specific files:
giggle search -i my_index -r chr1:1000000-2000000 -f "simple"Statistical analysis with query file:
giggle search -i my_index -q query.bed.gz -s
# Output includes: odds_ratio, Fisher's tests, combo_scoreGroup results by query interval:
giggle search -i my_index -q query.bed.gz -v -ocd test/unit && makeRequires bedtools in PATH.
# you may need to up the ulimit on some systems
# ulimit -n 1300
cd test/func && ./giggle_tests.sh# Download pre-built index
wget https://s3.amazonaws.com/layerlab/giggle/roadmap/roadmap_sort.tar.gz
tar -zxvf roadmap_sort.tar.gz
# Build index (if "Too many open files": ulimit -Sn 16384)
giggle index -s -f -i "roadmap_sort/*gz" -o roadmap_sort_b
# Download query data
wget ftp://ftp.ncbi.nlm.nih.gov/geo/samples/GSM1218nnn/GSM1218850/suppl/GSM1218850_MB135DMMD.peak.txt.gz
# Filter to top peaks
zcat GSM1218850_MB135DMMD.peak.txt.gz \
| awk '$8>100' \
| cut -f1,2,3 \
| bgzip -c > query.bed.gz
# Search with significance
giggle search -s -i roadmap_sort_b/ -q query.bed.gz > results.txtPre-built indexes available for download:
- Roadmap Epigenomics: https://s3.amazonaws.com/layerlab/giggle/roadmap/roadmap_sort.tar.gz
- UCSC Genome Browser: https://s3.amazonaws.com/layerlab/giggle/ucsc/ucscweb_sort.tar.gz
- FANTOM5: https://s3.amazonaws.com/layerlab/giggle/fantom/fantom_sort.tar.gz
Interactive heatmap: http://ryanlayer.github.io/giggle/
GIGGLE can run as a web service using libmicrohttpd.
# Install dependencies
mkdir -p $HOME/usr/local/
# libmicrohttpd
wget http://ftpmirror.gnu.org/libmicrohttpd/libmicrohttpd-0.9.46.tar.gz
tar zxvf libmicrohttpd-0.9.46.tar.gz
cd libmicrohttpd-0.9.46
./configure --prefix=$HOME/usr/local/ && make && make install
cd ..
# json-c
wget https://github.com/json-c/json-c/archive/json-c-0.12.1-20160607.tar.gz
tar xvf json-c-0.12.1-20160607.tar.gz
cd json-c-json-c-0.12.1-20160607
./configure --prefix=$HOME/usr/local/ && make && make install
cd ..
export LD_LIBRARY_PATH=$HOME/usr/local/lib/
# Build server
cd giggle
make serverRun servers:
giggle/bin/server_enrichment -i roadmap_sort_b/ -u /tmp/ \
-d giggle/examples/rme/data_def.json -p 8080 &
giggle/bin/server_enrichment -i ucscweb_sort_b/ -u /tmp/ \
-d giggle/examples/ucsc/data_def.json -p 8081 &
# Access at:
# http://ryanlayer.github.io/giggle/?primary_index=localhost:8080&ucsc_index=localhost:8081Note: These community-maintained bindings may be outdated.
python-giggle by Brent Pedersen
from giggle import Giggle
index = Giggle('my-index') # or Giggle.create('new-index', 'files/*.bed')
print(index.files)
result = index.query('chr1', 9999, 20000)
print(result.n_total_hits)
for hit in result[0]:
print(hit)Installation:
# Requires: zlib, libcurl, libcrypto, libbz2, liblzma
git clone --recursive https://github.com/brentp/python-giggle
cd python-giggle
python setup.py installgo-giggle by Brent Pedersen
import giggle "github.com/brentp/go-giggle"
index := giggle.Open("/path/to/index")
res := index.Query("1", 565657, 567999)
index.Files() // all files in index
res.TotalHits() // total count
res.Hits() // []uint32 hits per file
res.Of(0) // []string results from first filegiggle-docker by Ryuichi Kubo
docker run kubor/giggle-docker giggle -hgiggle-singularity by Hugo Guillen
giggle.sh check # verify configuration
giggle.sh pull # create container
giggle.sh index # create an index
giggle.sh search # search an index