EUVIMEDEUVIMED entdecken
Crossref

DeepIMB: Imputation of non-biological zero counts in microbiome data

Hanbyul Song, Md Mozaffar Hosain, Taesung Park iD

Genes & Genomics · 2026

DOI: 10.1007/s13258-025-01693-0 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Genes & Genomics, 2026. Verfasst von Hanbyul Song und 2 weiteren Autor:innen. Beteiligte Einrichtung: Interdisciplinary Program in Bioinformatics, Seoul National University, Seoul, 08826, Republic of Korea. Hauptthemen laut MEDLINE: Microbiota, Deep Learning. Die Arbeit nennt 42 Literaturangaben. Gefördert von: National Research Foundation of Korea.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND: The high prevalence of non-biological zero counts, arising from low sequencing depth and sampling variation, presents a significant challenge in microbiome data analysis. These zeros can distort taxon abundance distributions and hinder the identification of true biological signals, complicating downstream analyses. OBJECTIVE: To address the challenges of non-biological zeros in microbiome datasets, we propose DeepIMB, a deep learning-based imputation method for microbiome data, specifically designed to accurately identify and impute non-biological zero counts while preserving biological integrity. METHODS: DeepIMB operates in two main phases. First, it identifies non-biological zeros using a gamma-normal mixture model applied to the normalized, log-transformed taxon count matrix. Second, it imputes these zeros with a deep neural network model that integrates diverse sources of information, including taxon abundances, sample covariates, and phylogenetic distances, thereby learning complex, nonlinear relationships within microbiome data. RESULTS: By leveraging integrated information from multiple data types, DeepIMB accurately imputes non-biological zeros while preserving true biological signals. In our two simulation studies, DeepIMB outperformed existing imputation methods in terms of mean squared error, Pearson correlation coefficient, and Wasserstein distance. CONCLUSION: DeepIMB effectively addresses the challenges posed by non-biological zeros in microbiome data. By improving the quality of the data and the reliability of downstream analyses, DeepIMB represents a significant advancement in microbiome research methodologies.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

MicrobiotaDeep LearningHumansSoftwarePhylogenyNeural Networks, Computer

Stichwörter der Autor:innen

Deep learningDifferential abundanceImputationMicrobiomeZero inflation

Autor:innen und Institutionen

  • Interdisciplinary Program in Bioinformatics, Seoul National University, Seoul, 08826, Republic of Korea.
  • Department of Statistics, Seoul National University, Seoul, 08826, Republic of Korea.
  • Interdisciplinary Program in Bioinformatics, Seoul National University, Seoul, 08826, Republic of Korea. tspark@stats.snu.ac.kr.
  • Department of Statistics, Seoul National University, Seoul, 08826, Republic of Korea. tspark@stats.snu.ac.kr.

ORCID-Profile: Taesung Park

Förderung

  • National Research Foundation of Korea

Zitieren

APA

Song, H., Hosain, M. M., & Park, T. (2026). DeepIMB: Imputation of non-biological zero counts in microbiome data. Genes & Genomics. https://doi.org/10.1007/s13258-025-01693-0

Vancouver

Song H, Hosain MM, Park T. DeepIMB: Imputation of non-biological zero counts in microbiome data. Genes & Genomics. 2026. doi:10.1007/s13258-025-01693-0

BibTeX
@article{Song2026,
  author = {Song, Hanbyul and Hosain, Md Mozaffar and Park, Taesung},
  title = {DeepIMB: Imputation of non-biological zero counts in microbiome data},
  journal = {Genes & Genomics},
  year = {2026},
  doi = {10.1007/s13258-025-01693-0}
}
Kostenlos starten →