SORTMERNA(1)

💖 NAME 💖

okay so this is SORTMERNA(1), from the User Commands corner of the manual world — and what it does is: filtering, mapping and OTU-picking NGS reads. that's the whole vibe. 🧬✨

---

🌸 SYNOPSIS 🌸

sortmerna --ref db.fasta,db.idx --reads file.fa --aligned base_name_output [OPTIONS]

that's it, that's the shape of your command. you give it a ref, you give it reads, you give it somewhere to put the aligned stuff, and then [OPTIONS] is where you go absolutely feral. 💅

---

🦋 DESCRIPTION 🦋

SortMeRNA is a biological sequence analysis tool for filtering, mapping and OTU-picking NGS reads. the core algorithm is based on approximate seeds, which is the whole reason it can do fast and sensitive analyses of nucleotide sequences at the same time. no cap. the main application is filtering rRNA from metatranscriptomic data. additional applications include OTU-picking and taxonomy assignation, which you can get at through QIIME v1.9+ (http://qiime.org - v1.9.0-rc1).

so what actually goes in and what comes out: SortMeRNA takes as input a file of reads (fasta or fastq format) plus one or multiple rRNA database file(s), and it sorts apart rRNA reads and rejected reads into two files that you, yes you, specify. optionally it can also hand you high quality local alignments of the rRNA reads against the rRNA database. it works with Illumina, 454, Ion Torrent and PacBio data, and it can produce SAM and BLAST-like alignments. it's giving versatility. 😌

---

✨ OPTIONS ✨

🎀 MANDATORY OPTIONS 🎀

these are not optional, bestie. these are the bare minimum. 🫶

• --ref STRING,STRING — the FASTA reference file and the index file, comma-separated.

example:

--ref /path/to/file1.fasta,/path/to/index1

if you're passing multiple reference sequence files, separate them by ':'. like this:

--ref /path/f1.fasta,/path/index1:/path/f2.fasta,path/index2

• --reads STRING — your FASTA/FASTQ reads file.
• --aligned STRING — the filepath + base file name for the aligned reads (the appropriate extension gets added for you, don't worry about it 🎀).

---

💅 COMMON OPTIONS 💅

• --other STRING — filepath + base file name for the rejected reads (again, appropriate extension gets added).
• --fastx BOOL — output FASTA/FASTQ file (default: off, applies to aligned and/or rejected reads).
• --sam BOOL — output SAM alignment (default: off, applies to aligned reads only).
• --SQ BOOL — add SQ tags to the SAM file (default: off).
• --blast INT — output alignments in various Blast-like formats:
  1. 0 — pairwise
  2. 1 — tabular (Blast -m 8 format)
  3. 2 — tabular + column for CIGAR
  4. 3 — tabular + columns for CIGAR and query coverage
• --log BOOL — output overall statistic (default: off).
• --num_alignments INT — report the first INT alignments per read reaching E-value (default: -1; and heads up, --num_alignments 0 signifies that all alignments will be output).

or (default) 👇

• --best INT — report the INT best alignments per read reaching E-value (default: 1), by searching --min_lis INT candidate alignments (--best 0 signifies all candidate alignments will be searched).
• --min_lis INT — search all alignments having the first INT longest LIS (default: 2). LIS stands for Longest Increasing Subsequence, and it gets computed using seeds' positions to expand hits into longer matches prior to Smith-Waterman alignment. fancy, I know. 👑
• --print_all_reads — output null alignment strings for non-aligned reads (default: off), into SAM and/or BLAST tabular files.
• --paired_in BOOL — both paired-end reads go in the --aligned fasta/q file (default: off, interleaved reads only, see Section 4.2.4 of the User Manual).
• --paired_out BOOL — both paired-end reads go in the --other fasta/q file (default: off, interleaved reads only, see Section 4.2.4 of the User Manual).
• --match INT — SW score (a positive integer) for a match (default: 2).
• --mismatch INT — SW penalty (a negative integer) for a mismatch (default: -3).
• --gap_open INT — SW penalty (a positive integer) for introducing a gap (default: 5).
• --gap_ext INT — SW penalty (a positive integer) for extending a gap (default: 2).
• -N INT — SW penalty for ambiguous letters (the N's) (default: scored as --mismatch).
• -F BOOL — search only the forward strand (default: off).
• -R BOOL — search only the reverse-complementary strand (default: off).
• -a INT — number of threads to use (default: 1).
• -e DOUBLE — E-value threshold (default: 1).
• -m INT — INT Mbytes for loading the reads into memory (default: 1024, and the maximum -m INT is 5872).
• -v BOOL — verbose (default: off).

---

🌈 OTU PICKING OPTIONS 🌈

• --id DOUBLE — %id similarity threshold (the alignment must still pass the E-value threshold, default: 0.97).
• --coverage DOUBLE — %query coverage threshold (the alignment must still pass the E-value threshold, default: 0.97).
• --de_novo_otu BOOL — FASTA/FASTQ file for reads matching database < %id (set using --id) and < %cov (set using --coverage) (alignment must still pass the E-value threshold, default: off).
• --otu_map BOOL — output OTU map (this is the input to QIIME's make_otu_table.py, default: off).

---

🔥 ADVANCED OPTIONS 🔥

see the SortMeRNA user manual for more details 👀

• --passes INT — three intervals at which to place the seed on the read (L is the seed length set in indexdb_rna(1), default: L,L/2,3).
• --edges INT — number (or percent, if the INT is followed by a % sign) of nucleotides to add to each edge of the read prior to SW local alignment (default: 4).
• --num_seeds INT — number of seeds matched before searching for candidate LIS (default: 2).
• --full_search BOOL — search for all 0-error and 1-error seed matches in the index rather than stopping after finding a 0-error match (<1% gain in sensitivity with up to four-fold decrease in speed, default: off).
• --pid BOOL — add pid to output file names (default: off).
• -h BOOL — help.
• --version BOOL — SortMeRNA version number.

---

okay and that's the whole page, bestie!! you're gonna slay this filtering run, I can literally feel it. 🫶

this was sortmerna 2.0, August 2015 — mwah, iconic, no notes, go run your pipeline 💅✨🦋