Fig 1.
Schematic representation of the lovastatin BGC from Aspergillus terreus (lov).
In red the biosynthetic genes for SM production, in gold the further essential genes, and in blue the genes not involved in the biosynthetic pathway.
Table 1.
Fungal proteomes included in the empirically optimized database.
Fig 2.
Schematic representation of the workflow of FunOrder.
Table 2.
Empirically characterized biosynthetic gene clusters used as positive controls.
Fig 3.
Kernel density plot of the ICQ values for co-evolutionary linked enzymes of different control sets.
BioPath, protein sets of conserved biosynthetic pathways of the primary metabolism (S2 Table); random GCs, randomly assembled protein sets from 134 fungal proteomes (Table 1); BGCs, previously empirically characterized fungal BGCs (Table 2); sequential GCs, co-localized genes from random loci of different ascomycetes (S3 Table).
Fig 4.
A selection of the standard output of the FunOrder analysis of the lovastatin BGC (lov).
Score plots of the first two principal components from a PCA performed on the strict distance matrix (A) and on the combined distance matrix (B). The biosynthetic genes and the further essential genes are indicated in red and gold, respectively. Clusters in the PCA are indicated by the dashed boxes.
Fig 5.
Kernel density plots of the relative discovery rate of essential or biosynthetic genes in 30 tested fungal BGCs.
Table 3.
Statistical evaluation of the performance of FunOrder in detecting relevant genes in BGCs.