lncRNA编码能力预测

参考文献
Five coding potential prediction tools with different intrinsic sequence-related features (composition, structural properties and motifs) and divergent filtering steps (Weikard et al. 2017) including CPC2 (score > 0.5) (Kang et al. 2017), CNCI (score > 0) (Sun et al. 2013b), CPAT (score > 0.36) (Wang et al. 2013), PLEK (score > 0) (Li et al. 2014) and FEElnc (default parameters) (Wucher et al. 2017) were employed and the transcripts that were predicted as protein coding genes by at least three of above tools were removed.

conda使用方法

###CPC2
#http://cpc2.cbi.pku.edu.cn/download.php
source /your_soft_dir/biosoft/conda/etc/profile.d/conda.sh
conda activate python27 #use python 2.7 env

human_fa_linear=/data/fasta/Fed_Fasting_combined.human.isoforms.line.fa
data_dir=/data/fasta
input_dir=/data/fasta/total_novel
out_dir=/data/fasta/cpc2_results
cpc2_file=/your_soft_dir/Coding_potential/CPC2/CPC2-beta/bin/CPC2.py
python ${cpc2_file} -i /data/fasta/total_novel.fa -o ${out_dir}/total_novel.cpc2_results


###CNCI
#https://github.com/www-bioinfo-org/CNCI download and use directly
data_dir=/data/fasta
sof_dir=/your_soft_dir/Coding_potential/CNCI-master
out_dir=/data/fasta/cnci_results
python $sof_dir/CNCI.py -f $data_dir/total_novel.fa -o $out_dir/total_novel -m ve -p 4

###CPAT
#https://sourceforge.net/projects/rna-cpat/files/?source=navbar  website
#https://github.com/likelet/LncPipe/tree/master/bin/cpat_model cpat modules download
#https://blog.csdn.net/u013241595/article/details/101160741   pip3 aconda 
#conda install r-base==3.5.1 --force-reinstall
source /your_soft_dir/biosoft/conda/etc/profile.d/conda.sh
conda activate flair_env #use python 3.5 env
data_dir=/data/fasta
out_dir=/data/fasta/CPAT_results
modle_dir=/your_soft_dir/Coding_potential/LncPipe-master/bin/cpat_model
cpat.py -g /data/fasta/total_novel.fa   -d ${modle_dir}/Human_logitModel.RData \
-x ${modle_dir}/Human_Hexamer.tsv -o ${out_dir}/total_novel

###plek
#https://sourceforge.net/projects/plek/files/
source /your_soft_dir/biosoft/conda/etc/profile.d/conda.sh
conda activate python27 #python 2.7 env
plek_dir=/your_soft_dir/Coding_potential/PLEK.1.2/PLEK.py
out_dir=/data/fasta/plek_results

python ${plek_dir} -fasta /data/fasta/total_novel.fa  -out ${out_dir}/total_novel -thread 15 -minlength 20


###FEELnc
#https://github.com/tderrien/FEELnc
#use conda to install 
#cd ${conda_dir}
#find ./ -name FEELnc_codpot.pl in python27 env
#then follow the install guide to creat and source
#source deactivate   firstly
#then re   conda activate python27 then it works

source /your_soft_dir/biosoft/conda/etc/profile.d/conda.sh
#source activate /your_soft_dir/biosoft/conda/pkgs/feelnc-0.1.1-pl526_5
conda activate python27
cd /data/fasta/FEELnc_results
codpot_dir=/your_soft_dir/biosoft/conda/pkgs/feelnc-0.1.1-pl526_5/bin/FEELnc_codpot.pl
FEELnc_codpot.pl -i /data/fasta/total_novel.fa -a /your_soft_dir/index/ori_genomic/hg38/fasta/gencode.v33.pc_transcripts.fa --mode=shuffle -p 18