Comparison of differences in microbial compositions between negative controls and subject samples with varying analysis configurations.
10.4168/aard.2018.6.5.255
- Author:
Hyojung KIM
1
;
Sang Pyo LEE
;
Shin Myung KANG
;
Sung Yoon KANG
;
Sungwon JUNG
;
Sang Min LEE
Author Information
1. Department of Health Sciences and Technology, GAIHST, Gachon University, Incheon, Korea.
- Publication Type:Original Article
- Keywords:
Microbiota;
Metagenome;
Computational biology
- MeSH:
Classification;
Computational Biology;
Healthy Volunteers;
Metagenome;
Microbiota;
RNA, Ribosomal, 16S
- From:Allergy, Asthma & Respiratory Disease
2018;6(5):255-262
- CountryRepublic of Korea
- Language:Korean
-
Abstract:
PURPOSE: Identifying microbial communities with 16S ribosomal RNA (rRNA) gene sequencing is a popular approach in microbiome studies, and various software tools and data resources have been developed for microbial analysis. Our aim in this study is investigating various available software tools and reference sequence databases to compare their performance in differentiating subject samples and negative controls. METHODS: We collected 4 negative control samples using various acquisition protocols, and 2 respiratory samples were acquired from a healthy subject also with different acquisition protocols. Quantitative methods were used to compare the results of taxonomy compositions of these 6 samples by varying the configuration of analysis software tools and reference databases. RESULTS: The results of taxonomy assignments showed relatively little difference, regardless of pipeline configurations and reference databases. Nevertheless, the effect on the discrepancy was larger using different software configurations than using different reference databases. In recognizing different samples, the 4 negative controls were clearly separable from the 2 subject samples. Additionally, there is a tendency to differentiate samples from different acquisition protocols. CONCLUSION: Our results suggest little difference in microbial compositions between different software tools and reference databases, but certain configurations can improve the separability of samples. Changing software tools shows a greater impact on results than changing reference databases; thus, it is necessary to utilize appropriate configurations based on the objectives of studies.
