Dataset Information

NBC update: The addition of viral and fungal databases to the Naive Bayes classification tool.

ABSTRACT:

Background

Classifying the fungal and viral content of a sample is an important component of analyzing microbial communities in environmental media. Therefore, a method to classify any fragment from these organisms' DNA should be implemented.

Results

We update the näive Bayes classification (NBC) tool to classify reads originating from viral and fungal organisms. NBC classifies a fungal dataset similarly to Basic Local Alignment Search Tool (BLAST) and the Ribosomal Database Project (RDP) classifier. We also show NBC's similarities and differences to RDP on a fungal large subunit (LSU) ribosomal DNA dataset. For viruses in the training database, strain classification accuracy is 98%, while for those reads originating from sequences not in the database, the order-level accuracy is 78%, where order indicates the taxonomic level in the tree of life.

Conclusions

In addition to being competitive to other classifiers available, NBC has the potential to handle reads originating from any location in the genome. We recommend using the Bacteria/Archaea, Fungal, and Virus databases separately due to algorithmic biases towards long genomes. The tool is publicly available at: http://nbc.ece.drexel.edu.

SUBMITTER: Rosen GL

PROVIDER: S-EPMC3284397 | biostudies-literature | 2012 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

NBC update: The addition of viral and fungal databases to the Naïve Bayes classification tool.

Rosen Gail L GL Lim Tze Yee TY

BMC research notes 20120131

<h4>Background</h4>Classifying the fungal and viral content of a sample is an important component of analyzing microbial communities in environmental media. Therefore, a method to classify any fragment from these organisms' DNA should be implemented.<h4>Results</h4>We update the näive Bayes classification (NBC) tool to classify reads originating from viral and fungal organisms. NBC classifies a fungal dataset similarly to Basic Local Alignment Search Tool (BLAST) and the Ribosomal Database Proje ...[more]

PMID: 22293603

Similar Datasets

Project description:Verbal autopsy (VA) deals with post-mortem surveys about deaths, mostly in low and middle income countries, where the majority of deaths occur at home rather than a hospital, for retrospective assignment of causes of death (COD) and subsequently evidence-based health system strengthening. Automated algorithms for VA COD assignment have been developed and their performance has been assessed against physician and clinical diagnoses. Since the performance of automated classification methods remains low, we aimed to enhance the Naïve Bayes Classifier (NBC) algorithm to produce better ranked COD classifications on 26,766 deaths from four globally diverse VA datasets compared to some of the leading VA classification methods, namely Tariff, InterVA-4, InSilicoVA and NBC. We used a different strategy, by training multiple NBC algorithms using the one-against-all approach (OAA-NBC). To compare performance, we computed the cumulative cause-specific mortality fraction (CSMF) accuracies for population-level agreement from rank one to five COD classifications. To assess individual-level COD assignments, cumulative partially-chance corrected concordance (PCCC) and sensitivity was measured for up to five ranked classifications. Overall results show that OAA-NBC consistently assigns CODs that are the most alike physician and clinical COD assignments compared to some of the leading algorithms based on the cumulative CSMF accuracy, PCCC and sensitivity scores. The results demonstrate that our approach improves the performance of classification (sensitivity) by between 6% and 8% compared with other VA algorithms. Population-level agreements for OAA-NBC and NBC were found to be similar or higher than the other algorithms used in the experiments. Although OAA-NBC still requires improvement for individual-level COD assignment, the one-against-all approach improved its ability to assign CODs that more closely resemble physician or clinical COD classifications compared to some of the other leading VA classifiers.