MicroHDF: predicting host phenotypes with metagenomic data using a deep forest-based framework

Briefings in Bioinformatics, Journal Year: 2024, Volume and Issue: 25(6)

Published: Sept. 23, 2024

Abstract The gut microbiota plays a vital role in human health, and significant effort has been made to predict phenotypes, especially diseases, with the as promising indicator or predictor machine learning (ML) methods. However, accuracy is impacted by lot of factors when predicting host phenotypes metagenomic data, e.g. small sample size, class imbalance, high-dimensional features, etc. To address these challenges, we propose MicroHDF, an interpretable deep framework where cascade layers forest units designed for handling imbalance high dimensional features. experimental results show that performance MicroHDF competitive existing state-of-the-art methods on 13 publicly available datasets six different diseases. In particular, it performs best area under receiver operating characteristic curve 0.9182 ± 0.0098 0.9469 0.0076 inflammatory bowel disease (IBD) liver cirrhosis, respectively. Our also shows better robustness cross-study validation. Furthermore, applied two high-risk IBD autism spectrum disorder, case studies identify potential biomarkers. conclusion, our method provides effective reliable prediction phenotype discovers informative features biological insights.

Language: Английский

---

Microbiome alterations in autism spectrum disorder

Nature Microbiology, Journal Year: 2023, Volume and Issue: 8(9), P. 1615 - 1616

Published: Aug. 17, 2023

Language: Английский

---

Fecal microbial marker panel for aiding diagnosis of autism spectrum disorders

Gut Microbes, Journal Year: 2024, Volume and Issue: 16(1)

Published: Oct. 28, 2024

Accumulating evidence suggests that gut microbiota alterations influence brain function and could serve as diagnostic biomarkers therapeutic targets. The potential of using fecal signatures to aid autism spectrum disorder (ASD) detection is still not fully explored. Here, we assessed the different levels microbial markers (taxonomy genome) in distinguishing children with ASD from age gender-matched typically developing peers (

Language: Английский

---

Leveraging human microbiomes for disease prediction and treatment

Trends in Pharmacological Sciences, Journal Year: 2024, Volume and Issue: unknown

Published: Dec. 1, 2024

Language: Английский

---

Topic modeling for multi-omic integration in the human gut microbiome and implications for Autism

Scientific Reports, Journal Year: 2023, Volume and Issue: 13(1)

Published: July 13, 2023

While healthy gut microbiomes are critical to human health, pertinent microbial processes remain largely undefined, partially due differential bias among profiling techniques. By simultaneously integrating multiple methods, multi-omic analysis can define generalizable processes, and is especially useful in understanding complex conditions such as Autism. Challenges with heterogeneous data produced by methods be overcome using Latent Dirichlet Allocation (LDA), a promising natural language processing technique that identifies topics documents. In this study, we apply LDA (16S rRNA amplicon, shotgun metagenomic, metatranscriptomic, untargeted metabolomic profiling) from the stool of 81 children without We identify topics, or summarize phenomena occurring within communities. then subset samples topic distribution, metabolites, specifically neurotransmitter precursors fatty acid derivatives, differ significantly between clusters deemed "cross-omic topics", which hypothesize representative observable regardless method. Interpreting find each represents particular diet, heuristically label cross-omic as: healthy/general function, age-associated transcriptional regulation, opportunistic pathogenesis.

Language: Английский

---