BERIL Knowledge Wiki

Method

InterProScan

Well corroborated. AI-generated wiki from AI-conducted research reports. Partially reviewed by scientists. See About This Wiki.

Sources cited
4 project reports and 1 cross-project digest, listed in the margin
Relations
supports 3refines 5
Linked from
12 pages10 topics

What this entity is

Canonical name: InterProScan. 1

Known aliases: No aliases are specified in this source. 1

Stable external identifier: No stable external identifier is specified in this source. 1

InterProScan is a domain-annotation method used in joins among kbase-ke-pangenome, Bakta annotations, InterProScan domains, and kescience-alphafold AlphaFold records. 1 Its annotation tables should be joined to the pangenome at the gene-cluster level: eggnog_mapper_annotations.query_name, Bakta annotations, and InterProScan annotations join to gene_cluster.gene_cluster_id, not gene.gene_id. 2 This refines the earlier description by making the required join grain explicit. 2

Key facts

  • Of 132,531,501 total gene clusters, 111,035,431 (83.8%) had at least one InterProScan domain annotation. 1
  • Across the analysed gene clusters, the mean number of InterProScan domain hits was 7.5 and the mean number of distinct InterPro (IPR) families was 3.3. 1
  • Among 38,051,842 gene cluster–UniProt pairs, mean domain hits increased from 0.59 for AlphaFold MSA depth below 10 to 10.83 for MSA depth at least 10,000. 1
  • Across the same 38,051,842 pairs, mean distinct InterPro families increased from 0.059 for AlphaFold MSA depth below 10 to 4.601 for MSA depth at least 10,000. 1
  • The association between AlphaFold MSA depth and InterProScan domain-hit count was Spearman ρ = 0.7563 across the full 38,051,842-pair dataset. 1
  • InterProScan domain-annotation coverage of 83.8% exceeded the 29.3% AlphaFold bridge coverage in the report. 1
  • The report found that the monotone relationship between MSA depth and domain hits held within core, auxiliary non-singleton, and auxiliary+singleton pangenome classes, with core genes showing slightly higher domain richness per MSA bin than accessory genes at equivalent depth. 1

These results support using InterProScan domain and family counts as an annotation-richness measure in pangenome-integration, while the report cautions that the MSA-depth correlation was not subgroup-stratified and that the AlphaFold-linked subset was biased toward better-studied organisms. 1

In a separate pan-bacterial cofitness analysis, InterProScan supplied Gene Ontology (GO) annotations with 68% gene coverage, reported as 3.6× better than the old SEED annotations. 3 This supports the existing use of InterProScan as an annotation-richness measure and refines it by showing that the method also enables functional-enrichment analysis: InterProScan GO produced 35/3,193 significant tests at FDR < 0.05, compared with 0/280 for the old SEED/KEGG analysis. 3 The enrichment analysis identified flagellum-dependent motility, flagellum assembly, histidine biosynthesis, and tryptophan biosynthesis among recurring signals, although these may reflect shared dispensability under laboratory conditions rather than direct co-regulation. 3

The phage-defense analysis refines InterProScan’s role from a general annotation source to a practical Pfam-domain discovery source: its primary domain table contained 833-million rows, compared with 18.8-million rows in the Bakta Pfam table. 4 The same scale is a computational constraint: the InterProScan table is approximately 833M rows and requires key filters before joins, with results retained in Spark until the final small output. 2 Filtering InterProScan for analysis = 'Pfam' and a curated signature list of approximately 20 accessions, then joining to gene_cluster, returned ~500K rows in under 90 seconds. 4 The same workflow found 0 Cas1 PF01867 hits in bakta_pfam_domains versus ~25K in InterProScan, so the report used InterProScan for defense-marker detection. 4 InterProScan uses version-free Pfam accessions such as PF01867, whereas Bakta uses versioned accessions such as PF01867.29; cross-database joins therefore need to strip Bakta versions. 4 This supports the earlier finding that InterProScan provides broader annotation coverage, while qualifying cross-source comparisons because apparent differences can reflect database scale and accession formatting rather than biology. 4

The plant-microbiome analysis supports this broader-coverage interpretation but refines it with a direct audit: 12 of 22 marker Pfams were absent from bakta_pfam_domains, despite being abundant in interproscan_domains, and the ten non-zero Bakta Pfams returned only 10–35% of the InterProScan hit counts. 5 Versioned-Pfam matching with LIKE patterns recovered 19,364 hits across 7,962 species for ten queried marker Pfams. 5 InterProScan was therefore used for secretion-system and cell-wall-degrading-enzyme detection in the refined plant-marker pipeline; these biological assignments were unaffected by the Bakta-table gap. 5 The same study annotated all 50 retained plant-enriched gene families with eggNOG descriptions and InterProScan domains, while 48/50 had GO terms and 39/50 had MetaCyc pathways, supporting InterProScan’s utility in integrated functional annotation. 5

Because pangenome gene-cluster identifiers are species-specific, they must not be compared directly across species; cross-species analyses should instead use shared representations such as COG categories, KEGG orthologs, or Pfam domains. 2 This refines the interpretation of InterProScan-richness comparisons: domain annotations can provide a shared representation, but the comparison still depends on explicit cross-species normalization. 2

Sources cited

  1. AlphaFold MSA Depth as a Lens on the Bacterial Annotation Gapalphafold_msa_annotation · cited 17× · primary source · refines
  2. KBase Data Lakehouse: Common Pitfalls & Gotchaspitfalls · cited 6× · cross-project digest · refines · supports
  3. AMR Co-Fitness Support Networksamr_cofitness_networks · cited 5× · supports · refines
  4. Pan-Bacterial Anti-Phage Defense Arsenalphage_defense_arsenal · cited 7× · refines · supports
  5. Plant Microbiome Ecotypes — Compartment-Specific Functional Guilds and Their Genetic Architectureplant_microbiome_ecotypes · cited 5× · supports · refines