甲基化检测软件——DeepSignal

最新推荐文章于 2024-06-17 09:45:32 发布

随风而逝*

最新推荐文章于 2024-06-17 09:45:32 发布

阅读量1.7k

点赞数 2

文章标签：甲基化 deepsignal 生物信息

本文链接：https://blog.csdn.net/weixin_41869644/article/details/90230971

版权

1.deepsignal是基于Python3构建的。

conda create -n deepsignalenv python=3.6  #创建python版本为3.6的deepsignalenv的虚拟环境

conda activate deepsignalenv  #进入虚拟环境

git clone https://github.com/bioinfomaticsCSU/deepsignal.git  #从GitHub上下载deepsignal

cd deepsignal

pip install deepsignal  #安装deepsignal

pip install ont-tombo  #安装tombo

conda deactivate  #退出虚拟环境

2.软件参数。

deepsignal -h
usage: deepsignal [-h] {extract,call_mods,train} ...

detecting base modifications from Nanopore sequencing reads, deepsignal contains three modules: 
	deepsignal extract: extract features from corrected (tombo) fast5s for training or testing
	deepsignal call_mods: call modifications
	deepsignal train: train a model, need two independent datasets for training and validating

optional arguments:
  -h, --help            show this help message and exit

modules:
  {extract,call_mods,train}
                        deepsignal modules, use -h/--help for help

3. 示例。

1.re-squiggle

# cmd: tombo resquiggle $fast5_dir $reference_fa
tombo resquiggle fast5s.al GCF_000146045.2_R64_genomic.fna --processes 10 --corrected-group RawGenomeCorrected_001 --basecall-group Basecall_1D_000 --overwrite

2.extract features

deepsignal extract --fast5_dir fast5s.al/ --reference_path GCF_000146045.2_R64_genomic.fna --write_path fast5s.al.CpG.signal_features.17bases.rawsignals_360.tsv --corrected_group RawGenomeCorrected_001 --nproc 10

3.call modifications

# the CpGs are called by using the CpG model of HX1 R9.4 1D
deepsignal call_mods --input_path fast5s.al.CpG.signal_features.17bases.rawsignals_360.tsv --model_path model.CpG.R9.4_1D.human_hx1.bn17.sn360/bn_17.sn_360.epoch_7.ckpt --result_file fast5s.al.CpG.call_mods.tsv --nproc 10 --is_gpu no

deepsignal call_mods --input_path fast5s.al/ --model_path model.CpG.R9.4_1D.human_hx1.bn17.sn360/bn_17.sn_360.epoch_7.ckpt --result_file fast5s.al.CpG.call_mods.tsv --reference_path GCF_000146045.2_R64_genomic.fna --corrected_group RawGenomeCorrected_001 --nproc 10 --is_gpu no

python /path/to/deepsignal/scripts/call_modification_frequency.py --input_path fast5s.al.CpG.call_mods.tsv --result_file fast5s.al.CpG.call_mods.frequency.tsv --prob_cf 0

4. train

# need two independent datasets for training and validating
# use deepsignal train -h/--help for more details
deepsignal train --train_file /path/to/train_data/file --valid_file /path/to/valid_data/file --model_dir /dir/to/save/the/new/model

本文参照：https://github.com/bioinfomaticsCSU/deepsignal