The package is dedicated to structurally compare sequences of homologous genes from different genomes of a species. Briefly, the package has been developed to perform the following analyses. Our Wiki provides more detailed explanation.
- geneseq can extract gene sequence, gene annotation, other featured data in GTF/GFF and BED format, transcribed sequences, protein sequences, and gene sequences with transcribed sequences highlighted. The positions of gene annotation and other features are adjusted to the coordinates relative to the newly extracted sequences.
- homocomp implements homologous searching and visualization of the comparison. With a query gene sequence, homologous sequences can be identified in a genome. The homologous gene could be found if a related GTF is supplied. The alignment between two homologous regions are plotted.
- homomine search the best homolog in a targeted genome, identify and annotate polymorphisms, including SNP, INDEL, and structural variation (SV).
- homostack visualizes alignments among multiple homologous sequences.
- homograph visualizes alignments of homologous sequences, determines number of haplotypes, and views the haplotype graph. This module is still under development.
Usages of these modules can refer to Wiki.
The development of homotools was motivated by the tedious procedure for extracting sequences and related information of a gene from a reference genome and homologous genes from other genomes. Unix shell scripts, Perl, and R were combined for the development.
Shell scripts, Perl, and R were combined for the development. Both Perl and R are generally installed. If needed, please refer to Perl and R for installation guides. Java is required to run SNPEff. BLAST+, MURmer4, BEDtools are also required.
To run homomine, the multiple sequence aligner MAFFT is required.
To run homograph, cd-hit and one of the following aligners for multiple sequence alignment are required.
Clustal Omega
MUSCLE
MAFFT
All required packages can be installed through conda:
git clone https://github.com/liu3zhenlab/homotools.git
cd homotools
conda env create -f homotools.yml
# test run
cd data
sh homocomp_testrun.sh
Alternatively, separate packages can be installed separately with Conda.
conda create -n homotools
conda activate homotools
conda install -c anaconda perl openjdk
conda install -c r r-base r-knitr r-rmarkdown
conda install -c bioconda blast mummer4 bedtools cd-hit clustalo muscle mafft
# clone homotools to a local directory.
git clone https://github.com/liu3zhenlab/homotools.git
perl ./homotools/homocomp
# test run
cd data
sh homocomp_testrun.sh
Please report any bugs or suggestion on github or by email to Sanzhen Liu (liu3zhen@ksu.edu).
The package homotools is distributed under MIT licence.
0.3.4: modified homomine_report.Rmd
0.3.3: modified homomine_seqpair2var
0.3.2: updated homograph
0.3.1: simplified homomini
0.3.0: updated homograph outputs
0.2.9: added the parameter of colClasses="character" to utils/homomine_report.Rmd to avoid recognizing nucleotide T as a logic value
0.2.8: changed "echo -e" to Perl code to be more relialbe in different systems
0.2.7: fixed a bug in homomine to properly display SV tables
0.2.6: added homomine and updated homograph
0.2.5: changed "delta-filter -1" to "delta-filter -r" in the filter for uniquely mapped alignments
0.2.4: added the module homomine to compare homologs in two genomes and annotate polymorphisms
0.2.3: allowed an automatic choice for strand (plus or minus) in homocomp