| name | jackhmmer |
| description | Use when running iterative sequence-to-sequence HMMER searches to expand a protein family from one or a few seed sequences against a sequence database. |
| disable-model-invocation | true |
| user-invocable | true |
jackhmmer
Quick Start
- Command:
jackhmmer
- Local executable:
/home/vimalinx/miniforge3/envs/bio/bin/jackhmmer
- Version: HMMER 3.4
- Full reference: See
references/help.md
When To Use This Tool
- Start from one or a few representative protein sequences and iteratively expand the hit set across a sequence database.
- Detect more remote homologs than a one-pass
phmmer search can usually recover.
- Build an evolving alignment of accepted hits over multiple rounds before downstream model curation.
- Prefer
hmmsearch when you already have a profile HMM, and hmmscan when the target is an HMM database.
Common Patterns
jackhmmer \
--tblout hits.tbl \
--domtblout domains.tbl \
-N 5 \
--cpu 8 \
query.fa \
targets.fa
jackhmmer \
-N 3 \
-A accepted_hits.sto \
query.fa \
targets.fa
jackhmmer \
-N 5 \
--incE 1e-5 \
--incdomE 1e-5 \
query.fa \
targets.fa
Recommended Workflow
- Start from a representative protein FASTA query and a rewindable target sequence database in FASTA or another supported sequence format.
- Save
--tblout and --domtblout outputs on the first run so iteration effects are inspectable.
- Cap the number of rounds with
-N and set explicit inclusion thresholds if family drift would be costly.
- Review accepted hits and saved alignments before promoting the result into
hmmbuild or downstream annotation.
Guardrails
- In this workspace the binary currently fails to start because
libopenblas.so.0 is missing, so fix the runtime environment before expecting live execution.
jackhmmer takes positional arguments in query-then-database order; the target <seqdb> cannot be a non-rewindable stdin stream.
- This is a sequence-vs-sequence iterative searcher, not a profile-vs-sequence tool.
jackhmmer does not accept curated --cut_ga, --cut_nc, or --cut_tc threshold modes.