Figure: Slide culture of a streptomyces species. Reference and additional information: Wikipedia Streptomyces is the largest genus of Actinobacteria presenting a complex developmental life cycle[1,2]. Bacteria of these genus play a major role in the production of drugs with natural origin. More than 40% of all known antibiotics are produced by streptomyces spec. though many of these secondary metabolites are also used as chemotherapy or immunosuppressant drugs[3]. An enormous number of crucial data related to this genus have been collected up to the present but the exploitation of the chemical diversity of bioactive compounds produced by this genus is still essential for novel drug discovery[4].

  1. [1] Kämpfer P (2006). "The Family Streptomycetaceae, Part I: Taxonomy". The prokaryotes: a handbook on the biology of bacteria (Dworkin, M et al., eds.). Berlin: Springer. pp. 538–604.
  2. [2] Challis GL and Hopwood DA (2003) Synergy and contingency as driving forces for the evolution of multiple secondary metabolite production by streptomyces species. P Natl Acad Sci USA 100: 14555–14561. PubMed
  3. [3] Bérdy J (2005) Bioactive microbial metabolites. J Antibiot 58(1):1-26. PubMed
  4. [4] Genilloud O, González I, Salazar O, Martín J, Tormo JR, Vicente F (2011) Current approaches to exploit actinomycetes as a source of novel natural products. J Ind Microbiol Biotechnol 38(3):375-89. PubMed

Building process

The basis for StreptomeDB was initially built in spring of 2012 with a collection of drugs produced by streptomyces spec. To increase the data set we started a text and data mining project to build up a more comprehensive description of natural products of streptomyces.

Data sources

The information contained in StreptomeDB has been collected from literature and enhanced with data of two existing datasources (Novel Antibiotics DataBase and KNApSAcK). Extraction from literature was performed via text mining of thousands of articles from PubMed followed by manual curation using a newly developed in-house platform.

Most of the compounds in terms of names, synonyms, and structures were derived from PubChem. If compounds were not available in PubChem, compounds were carefully drawn and IUPAC names as well as the names of the articles where compounds first occurred were assigned as names and synonyms .

Scaffold browser

The shown scaffolds were extracted from the molecules in StreptomeDB with the Scaffold Decomposition tool included in Canvas 2.3. Level 0 scaffolds are herein root cyclic or policyclic indipendent structures. Two level 0 entities connected by an aliphatic or functionized linker define a level 1 scaffold, and so on. Those structures are sorted by their frequency of occurrence in all available structures in StreptomeDB.

Phylogenetic Tree

The phylogenetic tree has been built from the hypothetical most common 16S rRNA sequence for each available strain. This consensus sequence considers all available, but possibly slighly different sequncing attempts of the same high quality, as well as the usually six or more sequences that are encoded in a strains genome. It shows 340 strains, which are parent strains for nearly 1300 substrains. The tree is therefore comprising nearly two third of the data in StreptomeDB. Strains that have a common ancestor node within 0.07 were clusered with DendroPy and marked by background colors. The alignment has been performed with Clustalw, phylogenetic analysis with the MEGA software package, and the final editing with the ETE 2 toolkit.


All 1H and 13C NMR spectra were generated with the NMR predicter tool from the command-line platform cxcalc(Marvin, 2015, ChemAxon). The fragmentation patterns in mass spectra and the resulting chemical structures were predicted by the CFM-ID software for molecules with less than 35 heavy atoms in positive ESI-MS/MS mode at 10, 20, and 40 V.

StreptomeDB Team

Michael Becer, Andreas Bechthold, Denise Deubel, Kersten Döring, Astrid Erber, Anika Erxleben, Stephan Flemming, Björn Grüning, Stefan Günther, Anna Hähnlein, Dennis Klementz, Xavier Lucas, Johannes Mosch, Irene Santillana, Christian Senger, Kiran Telukunta, Oliver Thomas

Progress text
Message text