AutoCLARK

What does it do?

Use AutoCLARK to identify species represented in raw reads or draft genome assemblies. AutoCLARK runs CLARK, a metagenomic classification tool, and reports the species detected in each requested sample.

AutoCLARK is useful when the expected species is uncertain or when you want a taxonomic profile of a sample. It is not a dedicated contamination-detection workflow: use ConFindr when the primary question is whether raw sequencing reads are contaminated.

For background, see the CLARK website.

How do I use it?

Subject

In the Subject field, enter:

AutoCLARK

Spelling matters, but matching is not case-sensitive.

Description

The first line must identify the input type:

  • fastq — analyze raw reads;
  • fasta — analyze draft genome assemblies.

Enter one SEQID per subsequent line.

Raw-read request

fastq
2026-SEQ-0001
2026-SEQ-0002

Assembly request

fasta
2026-SEQ-0001
2026-SEQ-0002

Attachments

No attachment is required. AutoCLARK retrieves the sequence data associated with each requested SEQID according to the selected input type.

Optional parameters

The supplied documentation does not identify optional parameters for the Redmine AutoCLARK automator.

Example

See issue 12819 for an example AutoCLARK request.

Interpreting results

When AutoCLARK finishes, it uploads:

abundance.xlsx

The workbook reports the species detected in each requested sample and their estimated proportions.

Interpret low-abundance classifications cautiously. The existing workflow guidance notes that species reported below approximately 1–2% are often classification artifacts rather than organisms truly present in the sample. This is a practical interpretation guideline, not a universal biological threshold; review the result in the context of input quality, expected organisms, database composition, and supporting analyses.

AutoCLARK reports taxonomic classifications. A secondary species classification does not by itself establish that a sample is contaminated.

How long does it take?

AutoCLARK usually takes approximately 10–15 minutes per request. Requests containing many SEQIDs can take substantially longer.

What can go wrong?

A requested SEQID is unavailable

Symptom: The Redmine issue receives a warning identifying unavailable sequences.

Likely cause: AutoCLARK cannot locate the raw reads or draft assembly requested for that SEQID.

What to do: Verify the SEQID, confirm that the selected input type is available, and submit a corrected request.

The input type is missing or incorrect

Symptom: AutoCLARK cannot determine which sequence files to retrieve or analyze.

Likely cause: The first Description line is missing, misspelled, or inconsistent with the available data.

What to do: Use fastq for raw reads or fasta for draft assemblies.

A low-proportion species is overinterpreted

Symptom: A species reported at a very low proportion is treated as definitively present.

Likely cause: Low-abundance CLARK assignments can be classification artifacts.

What to do: Review the proportion, expected sample composition, sequence quality, and supporting evidence. Use ConFindr when contamination in raw reads is the specific question.

  • ConFindr — detects intra-species and inter-species contamination in raw sequencing reads.
  • Unknown Isolate — identifies an uncertain isolate from a draft genome assembly using rMLST, MASH, ANIb, and ANIm evidence.
  • Kraken2/Bracken and MetaPhlAn — provide metagenomic taxonomic analysis with different methods and trade-offs.