SVIM(1)

πŸ’– SVIM(1) β€” the structural variant caller that ate πŸ’–

okay bestie, sit down, hydrate, and let me tell you about svim, which lives in section 1 of the manual (user commands, obviously, it's a whole program) ✨

🌸 NAME 🌸

svim β€” a structural variant caller for long sequencing reads. yes. long reads. the main character of sequencing data πŸ’…


πŸ¦‹ SYNOPSIS πŸ¦‹

usage: svim [-h] [--version] {reads,alignment} ...

like, that's the whole shape of it. pick a mode, or ask for help, or ask for the version. that's the vibe check πŸŽ€


πŸ’œ DESCRIPTION πŸ’œ

SVIM β€” and yes, you pronounce it like "SWIM", not "ess-vim", I will not be taking questions β€” is a structural variant caller for long reads. it's out here discriminating between six different variant classes, no cap:

  • deletions
  • tandem duplications
  • interspersed duplications
  • inversions
  • insertions
  • translocations

and here's the part where SVIM is genuinely unique, like it's not even a competition: it can extract both the genomic origin and the genomic destination of duplications. both. origin AND destination. the way that it does that… obsessed πŸ˜­πŸ‘‘


✨ the four major steps ✨ (the pipeline, the journey, the character arc)

  1. COLLECT β€” detects signatures for SVs in long read alignments πŸ•΅οΈ
  2. CLUSTER β€” merges signatures that come from the same SV πŸ«‚
  3. COMBINE β€” combines clusters from different genomic regions and classifies them into distinct SV types 🧩
  4. GENOTYPE β€” uses alignments spanning SVs to determine their genotype 🧬

four steps. iconic behaviour. periodt.


🌈 the two ways to feed it 🌈

SVIM can process two types of input, bestie, so pick your fighter:

  • first up: it can detect SVs from raw reads by aligning them to a given reference genome first:
SVIM.py reads [options] working_dir reads genome
  • or alternatively: it can detect SVs from existing read alignments in SAM/BAM format:
SVIM.py alignment [options] working_dir bam_file

both are valid. both are slay. choose based on what you already have lying around πŸ’–


🌸 positional arguments 🌸

  • {reads,alignment} β€” these are the modes. the fork in the road. the two doors. choose one, you can't have both πŸ™ˆ
  • reads β€” Detect SVs from raw reads. Align reads to given reference genome first. (yes, it does the aligning for you first, it's giving generous ✨)
  • alignment β€” Detect SVs from an existing alignment. (you already did the aligning, so it just gets straight to the point πŸ’…)

πŸŽ€ optional arguments πŸŽ€

  • -h, --help β€” show this help message and exit. the classic. the comfort blanket. when in doubt, -h 🫢
  • --version, -v β€” show program's version number and exit. very demure, very mindful, tells you what version you're on 🌟
fr fr though β€” if you're ever lost, just hit -h. it will never judge you. it's literally there for that πŸ’œ

πŸ‘‘ AUTHOR πŸ‘‘

This manpage was written by Nilesh Patra for the Debian distribution and can be used for any other usage of the program. ty for your service Nilesh, we appreciate you 🫑🌸


🌈 SEE ALSO 🌈

okay so there aren't any SEE ALSO entries in this one, bestie β€” that's the whole page. no cross-references, no extra reading, we keep it self-contained. a minimalist queen πŸ’…


psst, bestie: remember that this is svim version 2.0.0, dated July 2021 β€” so if you're running something wildly different, don't come for me, come for the version number πŸ’€βœ¨

and that's the whole page! go call some structural variants, you absolute icon πŸ¦‹πŸ’–

πŸ“¦ where this page comes from