CentIER is a bioinformatics tool for identifying and annotating centromere regions of the genome that can be accurately identified.Instead of identifying centromere by identifying tandem repeats, CentIER was designed with sequence features including tandem repeats, retrotransposons, k-mer frequency distributions in mind, and allows users to further define centromere regions by entering a genome structure annotation file and a Hi-C signal file. While recognizing monocentric Chromosomes, we also designed a program to recognize the centromere regions of Metapolycentric Chromosomes.
By predicting the centromere of Arabidopsis, rice, corn, soybean, mulberry and other species, we found that CentIER maintained an accuracy of more than 90 percent. In maize, where centromere is difficult to predict, CentlER accurately predicted the centromere region of 9 out of 10 chromosomes.
The tool has the following software or package requirements.
apt-get install genometools when you are root user. Otherwise you only use make -j4 to Compile and install it. The gt requires cario and pango, you can use conda to install them. If you don't need it, use the cairo=no make parameter. In that case, run make cairo=no cleanup before running make cairo=no to start with a clean build environment. We recommend using gt_V1.6.5. pip install. Usually, You only need install pyfastx because ltr_retriever env have include other packages. You do not need to specify version number of packages. If some update case error, you can specify ltr_retriever=3.0.0, pyfastx=2.1.0, pandas=2.2.2, numpy=2.0.1, scipy=1.14.0.The following is an example of the installation procedure
# Create centier enviroment and install ltr_retriever
conda create -n centier -c bioconda -c conda-forge ltr_retriever=3.0.0
conda activate centier
pip install pyfastx==2.1.0
# install gt
# if you have not download gt, please use this link to get it.
wget https://github.com/genometools/genometools/archive/refs/tags/v1.6.5.tar.gz
tar -zxf genometools_v1.6.5.tar.gz
cd genometools_v1.6.5
make -j4
export PATH=$PATH:your_PATH/genometools_v1.6.5/bin
# install CentIER
# if you have a dictoinary name software to install software.
git clone https://github.com/simon19891216/CentIER.git
# Add the CentIER to PATH. you can add this code to .bashrc.
export PATH=$PATH:your_PATH/CentIER$ ./CentIERv3.0.py -h
usage: CentIERv3.0.py [-h] [-o OUTPUT] [--gff GFF] [-k KMER_SIZE]
[-c CENTER_TOLERANCE] [--step_len STEP_LEN]
[--mul_cents] [--matrix1 MATRIX1]
[--matrix2 MATRIX2] [--bed1 BED1] [--bed2 BED2]
[--MINGAP MINGAP]
[--SIGNAL_THRESHOLD SIGNAL_THRESHOLD]
genome
positional arguments:
genome Genome fasta file need to annotation
options:
-h, --help show this help message and exit
-o OUTPUT, --output OUTPUT
output path
--gff GFF optional, annotation file (gff or gtf format)
-k KMER_SIZE, --kmer_size KMER_SIZE
the size of kmer
-c CENTER_TOLERANCE, --center_tolerance CENTER_TOLERANCE
the fine-tuning size between two regions
--step_len STEP_LEN the size between two regions
--mul_cents This parameter is used to retain all potential
centromeric regions, especially when the
chromosomes of the species under study may not
have a single, defined centromere; it should be
included in such cases.
--matrix1 MATRIX1 Path to matrix of 100000 file
--matrix2 MATRIX2 Path to matrix of 200000 file
--bed1 BED1 Path to bed file of matrix1
--bed2 BED2 Path to bed file of matrix2
--MINGAP MINGAP Minimum gap value n*100000 (default: 2)
--SIGNAL_THRESHOLD SIGNAL_THRESHOLD
Signal threshold value (default: 0.7)CentIER have one positional argument genome.fasta. If you only have a T2T assembly (T2T level or Chromosome level genomes), you can running this tools as follow:
CentIER.py genome.fastaIf you have gff3 annotation file, you can running program as follow:
CentIER.py --gff genome.gff genome.fastaIf you have Hi-C data, you can use Hi-C matrix to identify regions of centremers more accuracy. The Hi-C matrix can be obtaion from Hi-Cpro. You can use the the Hi-C matrix as follow:
CentIERv3.0.py --matrix1 sample.100000.matrix --matrix2 sample.20000.matrix --bed1 sample.100000_abs.bed --bed2 sample.20000_abs.bed genome.fastaFor more detailed use, or if you just want to locate the centromere region through Hi-C, you can refer to CentromFind repository.
Depend on our test result, the BUG that ids will not be match if your Chr_ID of genome fasta does not follow the form of the following case.
# fasta sample. The sequence id must be "ChrN". if use Chr_1, the program will error.
>Chr1
AAACCCTAAACCCTAAACCCTAAACCCTAAACCCTAAACC
>Chr2
CTGGTTGAAAATCATTGTGTATATAATGATAAT
>Chr3
GCTACGATCTACATTTGGGAATGTGAGTCTCTTATTGT
>Chr4
ACCCTAAACCCTAAACCCTAAACCCTAA
>Chr5
CCTAAACCCTAAACCCTAAACCCTAAACCCTA
>Chr6
TCTTATTAATTGTTTGGACTGTTTATGTTTGGACATT
......The output_path of CentIER can be specifed by -o and if you don't use -o, the default output path is ./CentIER_final_results. The output file list as follow:
$ ls
ColCEN.fasta_all_centromere_seq.txt
ColCEN.fasta_all_centromere_seq.txt_l1.txt
ColCEN.fasta_centromere_range.txt
ColCEN.fasta_draw_cen.svg
ColCEN.fastaLTR-hmmresult.txt
ColCEN.fasta_LTR_positionb.txt
ColCEN.fasta_ltr_position.txt
ColCEN.fasta_ltr_position.txt_seqt.txt
ColCEN.fasta_ltr_position.txt_seq.txt
ColCEN.fasta_LTR_statistics.txt
ColCEN.fasta_monomer_in_centromere.txt
ColCEN.fasta_monomer_seq.txtThe program will produce many tmp file with start with basename of your genome file in current path (./your_genome.fa.*). After running, you can remove them manually.
Wait for more information
if you use CentIER or CentromFind,
please cite this atricle:
CentIER: accurate centromere
identification for plant genome
Xu, D. et al. CentIER: accurate centromere identification for plant genome. Plant Communications 101046 (2024) doi:10.1016/j.xplc.2024.101046.
If you have any problem while running the program, you can contact us through the ways as follow:
You can submit issues in this repository too. Thanks for your using our program.
Author: Xu dong, Yang jinbao