Customization of a DADA2-based pipeline for fungal internal transcribed spacer 1 (ITS1) amplicon data sets

Abstract

Identification and analysis of fungal communities commonly rely on internal transcribed spacer-based (ITS-based) amplicon sequencing. There is no gold standard used to infer and classify fungal constituents since methodologies have been adapted from analyses of bacterial communities. To achieve high-resolution inference of fungal constituents, we customized a DADA2-based pipeline using a mix of 11 medically relevant fungi. While DADA2 allowed the discrimination of ITS1 sequences differing by single nucleotides, quality filtering, sequencing bias, and database selection were identified as key variables determining the accuracy of sample inference. Due to species-specific differences in sequencing quality, default filtering settings removed most reads that originated from Aspergillus species, Saccharomyces cerevisiae, and Candida glabrata. By fine-tuning the quality filtering process, we achieved an improved representation of the fungal communities. By adapting a wobble nucleotide in the ITS1 forward primer region, we further increased the yield of S. cerevisiae and C. glabrata sequences. Finally, we showed that a BLAST-based algorithm based on the UNITE+INSD or the NCBI NT database achieved a higher reliability in species-level taxonomic annotation compared with the naive Bayesian classifier implemented in DADA2. These steps optimized a robust fungal ITS1 sequencing pipeline that, in most instances, enabled species-level assignment of community members.

Keywords: Fungal infections; Infectious disease; Microbiology.

Figure 1. Overview of data set.

(A) Sample and Sequencing workflow. (B) Length variation in the amplified ITS1 region of strains within the mock community. The 2 S. cerevisiae ASVs differed by one nucleotide in length. (C) Pairwise Levenshtein distance between the expected ITS1 amplicon sequences included in the mock community. (D) Nucleotide differences between expected highly similar ITS1 amplicon sequences.

Figure 2. Performance of DADA2 on the mock community data set.

(A) Strain resolution of DADA2 (ASV) compared with UPARSE (OTU). The balanced community has equal 18S rDNA copy number normalized amounts of DNA per strain. The extreme 1 community include equal 18S rDNA copy number normalized amounts of DNA per strain, except for A. fumigatus and M. guilliermondii, which were included at 50-fold dilution. The extreme 2 community include equal 18S rDNA copy number normalized amounts of DNA per strain, except for A. fischeri and M. caribbica, which were included at 50-fold dilution. (B) Representative quality profile of raw reads that were denoised into exact sequence matches to A. fumigatus and C. albicans. The line represents the median Phred score at that position, while the shaded area represents the 25th to 75th percentiles. (C) Impact of varying truncQ and maxEE on the number of species-specific reads.

Figure 3. Effect of customizing filtering variables of mock community data set.

(A and B) Number of reads and ration of expected to nonexpected sequences for different combinations of truncQ and maxEE. (C) Comparison of the Levenshtein distance for all nonexpected sequences to the nearest expected ASV reference sequences.

Figure 4. Effect of customizing filtering values on patient samples.

(A) Taxonomic composition of fecal samples according to the filtering strategy used. The arrow shows the retention of reads from Aspergillus species, which were completely discarded by the standard filtering strategy. (B) Number of reads retained by DADA2 according to filtering strategy used. (C) Phred quality scores along the R1 and R2 reads of selected fungal genera, generated from ref. 6. (D) Taxonomic composition of representative samples from ref. 6 according to the filtering strategy used.

Figure 5. Length-specific biases in fungal ITS1 amplicon sequencing and adaptation of forward primers for better recall of S. cerevisiae and C. glabrata.

(A) Single nucleotide difference between the ITS1 forward primer and the S. cerevisiae reference genome. (B) Impact on the relative abundance of S. cerevisiae and C. glabrata when using a wobble forward primer allowing for the single nucleotide difference between the ITS1 forward primer and the S. cerevisiae reference genome.

Figure 6. Percentage of correct species-level annotation for different algorithm/database combinations.