目录
- .gitignorelog progress of writing sequencing2个月前
- BUSCO_phylogenomics.pyadd option to choose min number of species for gene tree construction. rename gene tree output directory structure. fix typos.8天前
- LICENSEa parameter to work in nucleotide space (--nt)2年前
- README.mdadd option to choose min number of species for gene tree construction. rename gene tree output directory structure. fix typos.8天前
- conda_build_package.shCreate conda_build_package.sh1年前
- conda_env.yamlSwitch to MAFFT (in auto mode) as default for multiple sequence alignment. Keep MUSCLE as an option (--muscle). Recent testing, MUSCLE was much slower for larger datasets with more species/BUSCO. Update README.1个月前
- count_buscos.pyprint sample name in count_buscos. output2年前
- pipeline.pngupdate pipeline.png image1个月前
邀请码
版权所有：中国计算机学会技术支持：开源发展技术委员会
京ICP备13000930号-9
京公网安备 11010802032778号
BUSCO_Phylogenomics
Jamie McGowan, 2026
This is a simple, easy-to-use pipeline to construct species phylogenies using BUSCO proteins. It works directly from BUSCO output and can generate concatenated supermatrix alignments and also gene trees of BUSCO families.
The pipeline identifies BUSCO proteins that are complete and single-copy in all input samples. Alternatively, you can account for missing data and choose to include BUSCO proteins that are complete and single-copy in a certain percentage of input samples. Each BUSCO family is individually aligned, trimmed, and then concatenated together to generate a supermatrix alignment. The pipeline also generates individual phylogenies for BUSCOs that are complete and single-copy.
🆘❓📧 Got a question or need some help? Open an issue or feel free to email me - jamie.mcgowan@ucd.ie
Dependencies
The pipeline requires the following dependencies:
These should be available from your
$PATH.You can install the BUSCO_Phylogenomics package with Conda from the Bioconda channel:
Alternatively, you can manually install the packages and dependencies using conda with the provided yaml file
conda_env.yaml, which will create a conda environment called BUSCO_phylogenomicsUsage
You should move all of your completed BUSCO output directories into the same directory.
Example usage:
This will look in the “BUSCO_results” directory for completed BUSCO runs, generate multiple sequence alignments for all complete single-copy proteins that were found in all samples, trim alignments with trimal and then concatenate them together, generating a concatenated alignment in Fasta and Phylip format along with a partitions file in NEXUS format. It will also generate gene trees for all BUSCO proteins that are complete and single-copy in at least 4 samples. The output will be stored in a directory named “output_busco_phylogenomics”. The pipeline is written to be executed on a single node/machine, here 8 parallel alignment/trimming/phylogeny jobs would run.
By default, MAFFT (–auto) is used for sequence alignment. Alternatively, you can use MUSCLE by specifying
--muscle.If you don’t want to generate gene trees, you can use the parameter
--supermatrix-onlyto only generate the concatenated alignment.If you don’t want to generate a concatenated alignment, you can use the parameter
--gene_trees_onlyto only generate gene trees.By default, the pipeline works in protein space (i.e., aligns amino acid sequences). The
🚨 WARNING:
--ntflag switches to using BUSCO nucleotide sequences instead of proteins.--ntnucleotide mode does NOT work if you used miniprot to find BUSCOs. This is because running BUSCO with miniprot only outputs amino acid sequences.If you have a patchy dataset and want to include BUSCO proteins in your concatenated alignment that aren’t universally present, you can use the
--percent_single_copyparameter.For example:
will include all BUSCO families that are complete and single-copy in at least 70% of samples in your concatenated alignment. Missing data will be represented by “?” characters in the concatenated alignment by default. You can specify a different character to represent missing data with the
--missing_characterparameter.The provided
count_buscos.pyscript can be used to count single-copy BUSCOs and summarise BUSCO presence/absences across samples to determine an appropriate cut-off for how much missing data to allow (–percent_single_copy).This will report how many BUSCOs are complete and single-copy in what percentage of samples and print a presence/absence table for each BUSCO family.
If you used BUSCO version 3 you should use the flag
--busco_version_3as the output structure of this version of BUSCO is slightly different to that of versions 4 and 5.Publications that use the BUSCO_phylogenomics pipeline (to Feb 2026; N = 105):