funbarRF: DNA barcode-based fungal species prediction using multiclass Random Forest supervised learning model

Abstract

BackgroundIdentification of unknown fungal species aids to the conservation of fungal diversity. As many fungal species cannot be cultured, morphological identification of those species is almost impossible. But, DNA barcoding technique can be employed for identification of such species. For fungal taxonomy prediction, the ITS (internal transcribed spacer) region of rDNA (ribosomal DNA) is used as barcode. Though the computational prediction of fungal species has become feasible with the availability of huge volume of barcode sequences in public domain, prediction of fungal species is challenging due to high degree of variability among ITS regions within species.ResultsA Random Forest (RF)-based predictor was built for identification of unknown fungal species. The reference and query sequences were mapped onto numeric features based on gapped base pair compositions, and then used as training and test sets respectively for prediction of fungal species using RF. More than 85% accuracy was found when 4 sequences per species in the reference set were utilized; whereas it was seen to be stabilized at ~88% if ≥7 sequence per species in the reference set were used for training of the model. The proposed model achieved comparable accuracy, while evaluated against existing methods through cross-validation procedure. The proposed model also outperformed several existing models used for identification of different species other than fungi.ConclusionsAn online prediction server “funbarRF” is established at http://cabgrid.res.in:8080/funbarrf/ for fungal species identification. Besides, an R-package funbarRF (https://cran.r-project.org/web/packages/funbarRF/) is also available for prediction using high throughput sequence data. The effort put in this work will certainly supplement the future endeavors in the direction of fungal taxonomy assignments based on DNA barcode.

Unambiguous identification of fungi: where do we stand and how accurate and precise is fungal DNA barcoding?

“…However, automated verification can be achieved through phylogeny-based analysis of metabarcoding reads that compute statistical support values for alternative placements. This can be achieved either through local alignments of BLAST hits under a Bayesian framework (Munch et al 2008;Porter and Golding 2011), with a probabilistic approach such as PROTAX Fungi (Abarenkov et al 2018), through a "random forest" learning tool (Meher et al 2019), or through read placement into a separately established reference tree (Berger et al 2011;Matsen et al 2012;Barbera et al 2019).…”

“…While the Bayesian framework in pplacer offers direct assessment of statistical confidence, the EPA allows the computing of bootstrap support values for potential alternative read placements. These options provide an automated, quantitative verification step not available through OTU clustering or BLAST mapping, except with approaches such as PROTAX Fungi and "random forest" learning (Abarenkov et al 2018;Meher et al 2019). Optionally, prior to invoking the EPA, the phylogenetic pattern of the metabarcoding marker over the fixed reference alignment can be analyzed using a maximum parsimony or maximum likelihood approach in order to compute a weight vector.…”

Efficacy and accuracy responses of DNA mini-barcodes in species identification under a supervised machine learning approach

“…Simple-logistic, IBK, PART, Attribute-selected Classifier, Bagging approaches were also implemented in another study [49] in this regard. SMO, BP-NN [50], RF [51], k-mer based approaches [52] [53] can also be perceived in recent studies. Naïve Bayes is also applied on COI [54] barcode database demonstrating misclassification rates and also on ribosomal databases [55] effectively.…”

Its2vec: Fungal Species Identification Using Sequence Embedding and Random Forest Classification

“…Several rRNA genes have been success-fully employed for fungal species identification, including the small ribosomal subunit, the large ribosomal subunit, the RNA polymerase II binding protein, and the internal transcribed spacer (ITS). Among these, the ITS (including ITS1 and ITS2 separated by the 5.8S genic region) has been widely adopted as a marker for fungal identification and diversity exploration [15][16][17][18][19] because this region is ubiquitous and shows great variation in sequence and length [9].…”

“…The Warcup training set was developed from the UNITE database and includes only sequences with authoritative taxonomic or lineage information [9]. In addition, ITS barcodes in the BOLD database (http://www.boldsystems.org/) [22] and the ITS1 database comprising sequences of NCBI GenBank (http:// www.ncbi.nlm.nih.gov/) have been used for fungal species identification [10,15]. DNA barcode-based taxonomic assignment can be achieved by using similarity-based or prediction-based (alignment-free) methods.…”