tree

Place bins in the reference genome tree.

This is the first step of the lineage-specific workflow. Genes are called in each bin with Prodigal (unless --genes is used), phylogenetically informative marker genes are identified with HMMER, and each bin is placed in the reference genome tree with gPlace.

--ali: generate a HMMER alignment file for each bin in <output folder>/bins/<bin id>/hmmer.tree.ali.txt.gz.
--nt: generate nucleotide gene sequences for each bin in <output folder>/bins/<bin id>/genes.fna.gz.
--genes: bins contain genes as amino acids instead of nucleotide contigs.
-t: number of threads, which is also the number of threads used by gPlace.
--gplace_max_memory: memory gPlace may use to place bins, e.g. 64G or 500M (a plain number is GiB). If the placement needs more memory, gPlace places bins tile by tile within this limit, which gives identical results. By default, gPlace sets this from the available memory. If the limit is too small, CheckM reports gPlace’s error message, which gives the smallest limit that can be used (see System requirements).

Called genes are written to <output folder>/bins/<bin id>/genes.*.gz (see Output files). Bins smaller than 100 kbp are processed with Prodigal’s metagenomic mode (-p meta).

Note

CheckM <= 1.2.10 placed bins with pplacer and provided the --reduced_tree and --pplacer_threads options, which were removed in CheckM 1.3.0.

Note

The following heuristic is used to establish the translation table used by Prodigal: use table 11 unless the coding density using table 4 is more than 5 percentage points higher than when using table 11 and the coding density under table 4 is > 70%. Distinguishing between tables 4 and 25 is challenging so CheckM does not attempt to distinguish between these two tables. If you know the correct translation table for your genomes, it is recommended that you call genes outside of CheckM and provide CheckM with the protein sequences (see --genes).

Arguments

usage: checkm tree [-h] [--ali] [--nt] [-g] [-x EXTENSION] [-t THREADS]
                   [--gplace_max_memory MEMORY] [-q] [--tmpdir TMPDIR]
                   bin_input output_dir

Place bins in the genome tree.

positional arguments:
  bin_input             directory containing bins (fasta format) or path to file describing genomes/genes - tab separated in 2 or 3 columns [genome ID, genome fna, genome translation file (pep)]
  output_dir            directory to write output files

options:
  -h, --help            show this help message and exit
  --ali                 generate HMMER alignment file for each bin
  --nt                  generate nucleotide gene sequences for each bin
  -g, --genes           bins contain genes as amino acids instead of nucleotide contigs
  -x, --extension EXTENSION
                        extension of bins (other files in directory are ignored) (default: fna)
  -t, --threads THREADS
                        number of threads (default: 1)
  --gplace_max_memory MEMORY
                        memory gPlace may use to place bins, e.g. 64G or 500M (a plain number is GiB) [default: set by gPlace from the available memory]
  -q, --quiet           suppress console output
  --tmpdir TMPDIR       specify an alternative directory for temporary files

Example: checkm tree ./bins ./output

Example

checkm tree -t 8 -x fa ./bins ./output
checkm tree -t 8 -x fa --gplace_max_memory 8G ./bins ./output