-
Notifications
You must be signed in to change notification settings - Fork 0
License
shiying-sxu/Attdeepcaller
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
# Attdeepcaller # create and activate an environment named attdeepcaller conda create -n attdeepcaller python=3.9.0 -y source activate attdeepcaller # install pypy and packages in the environemnt conda install -c conda-forge pypy3.6 -y pypy3 -m ensurepip pypy3 -m pip install mpmath==1.2.1 # install python packages in environment pip3 install tensorflow pip3 install tensorflow-addons tables conda install -c anaconda pigz==2.4 cffi==1.14.4 -y conda install -c conda-forge parallel=20191122 zstd=1.4.4 -y conda install -c conda-forge -c bioconda samtools=1.10 -y conda install -c conda-forge -c bioconda whatshap=1.4 -y conda install -c conda-forge xz zlib bzip2 automake curl -y conda install seaborn #Go to the installation location of the Attdeepcaller program (download to the specified location and extract the samtools and longphase packages) Cd Attdeepcaller #Install libclair3: make PREFIX=${CONDA_PREFIX} Train and test the attdeepcaller model: 1.Data preparation conda activate attdeepcaller The output union.vcf.gz is placed in the specified folder: OUTPUT_DIR ①sh subsampledata.sh#Downsampled data ②sh 0_rep_uni.sh #Normalized data 2.pileup data training ④sh 1_trainpileupmodel.sh #Training preparation(pileup) ⑤sh 2_pileup_training.sh #Training(pileup) 3.full-alignment training ①sh 3_trainfullalignment_ont_r1.sh #Training preparation(full-alignment) ②sh 4_training_fullalignment_ont_r1.sh#Training(fullalignment) 4.Testing ①sh 5_attdeepcaller_ont_quick_demo-HG002.sh #Testing(conda activate attdeepcaller). conda activate happy-env ②sh 6_visualization_attdeepcaller_ont_quick_demo-HG002.sh #Test visualization(conda activate happy-env) conda activate attdeepcaller ③python GetOverallMetrics.py --happy_vcf_fn=/work/Clair3-main-sy/clair3_ont_quickDemo/output/happy.vcf.gz --output_fn=/work/Clair3-main-sy/clair3_ont_quickDemo/output/metrics ##Statistical result 5.Model available The trained model is available at the following link: Link: https://pan.baidu.com/s/1NqcE5xtmZS-SFb5Md2o4dQ?pwd=2024 Extraction code: 2024 6. Data available Reference genomes GRCh38_no_alt https://ftp.ncbi.nlm.nih.gov/genomes/all/GCA/000/001/405/GCA_000001405.15_GRCh38/seqs_for_alignment_pipelines.ucsc_ids/GCA_000001405.15_GRCh38_no_alt_analysis_set.fna.gz 7.Illumina data preprocessing illumina need to install realigner Installation package # Install boost library conda install -c conda-forge boost=1.67.0 -y #echo "Environment:" ${CONDA_PREFIX} echo "Environment:" /home/user/anaconda3/envs/attdeepcaller # Make sure in Attdeepcaller directory cd Attdeepcaller cd preprocess/realign g++ -std=c++14 -O1 -shared -fPIC -o realigner ssw_cpp.cpp ssw.c realigner.cpp g++ -std=c++11 -shared -fPIC -o debruijn_graph -O3 debruijn_graph.cpp -I /home/user/anaconda3/envs/attdeepcaller/include -L /home/user/anaconda3/envs/attdeepcaller/lib View in Supplementary Materials.
About
No description, website, or topics provided.
Resources
License
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published