v3.2.4

• Linting.
• Add output blastout.filtered.txt and extracted.filtered.fa.
• Skip invalid samples in stage_one instead of raise exceptions.
• Fix a bug causing 16S calculation being slow and slightly overestimated.
• Fix a bug about TPM normalization.

v3.2.3

• Add a gene capO
• Reduce peak meamory usage for large files #29.

v3.2.2

• Fix naming issue #15.

v3.2.1

• Fix fastq reading bug.
• Rename scov to copy in output files.

v3.2

• Simplify interface.
• Remove dependencies on manually entered metadata.txt.
• Support bioconda installation.

v3.1.4

• Remove a duplicated gene ID (ARCH67_P638154538) in KO30 name list.
• Remove gene msrE, mphE and some multidrug genes from database.
• Add RPKM/TPM normalization.

v3.1.3

• Update database to the release version (29082022 short).
• Add parameter -s for skipping 16s/cells calculation in stageone.
• Fix a bug when empty _2 file is being used as input.

v3.1.2

• Support customized database (testing).
• Fix a bug of ppm normalization, now the formula is #hits * 1e6 / #reads.
• Simply stagetwo's script, add more information for users.

v3.1.1

• Fix a bug related to the calculation of 16S rDNA copies. Now the numerator is the aligned length instead of read length, and the denominator is the subject length instead of 1432 (the averaged length of 16S rDNA). This will lead to a slight drop of #16S in meta_data_online.txt.
• Fix a bug when estimating 16S rDNA numbers, which causes some 16S to be counted more than one time when multiple high scoring pairs (HSPs) are returned by blastn.

v3.1

• Minor changes to the SARG database (see SARG v3.0-M).
• Update diamond to the latest version (from 0.9.24 to 2.0.15), add a new parameter -w (query coverage, default 65%) and modify the default parameter of -v (identity, default 40%) in stageone to compensate the difference in USCMG estimation.
• Add a mt_mode switcher in stagetwo's blastx to make the pipeline faster when more than 5 cores are available.
• Add a logger to make stagetwo's stdout more clear.