The number of microbiome-based or related studies has increased dramatically resulting in the availability of data on human microbiome composition and function for different areas of problems. Such studies provide the raw data to deeply explore host-microbiome associations and their relation to the development and progression of various complex diseases, environmental phenomena, and abilities or features. Advanced data analysis approaches are needed for the comprehensive utilization and investigation of all data available from these studies, keeping in mind the intrinsic nature of microbiome data when it comes to composition and functional capabilities. Taxonomy-aware approaches and analysis pipelines can potentially help detect the hidden associations and patterns between microbiome and case states, such as for a disease, in order to develop customized prediction models.
The development of prediction models that are customized or tailored to solve a certain problem can be realized via Machine Learning (ML). ML algorithms allow for the development of classification models that can predict a disease state after being trained on a set of informative biomarkers from the microbial communities used to represent each sample of a known group. In order to leverage ML into more translational research related to the microbiome and strengthen our ability to extract meaningful and useful patterns, it is important for the models’ results to be interpretable. The feature space that is built of informative features or biomarkers can be leveraged for this purpose. The investigation should not be limited to only the composition of the space of biomarkers, but also how they act or contribute to the observed phenotypes. This opens up the possibility to incorporate the functional component into the space of features used for training and observation interpretation. Advances in ML algorithms and preprocessing methods, such as feature selection and engineering, allow for analysis of the microbiome that leads to interpretable outcomes and phenotypes, which can be utilized for the development of microbe-targeted solutions, such as therapies and ultimately personalized medicine.
Topics of interest include the following:
1. Knowledge representation and reasoning of microbial data
2. Applications of machine learning in microbiome studies
3. Feature selection and engineering for biomarker identification
4. Microbiome analysis for disease prediction
5. Microbial data analysis for predicting gene functionality
Keywords:
microbiome analysis, gene functionality, machine learning, microbial data
Important Note:
All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
The number of microbiome-based or related studies has increased dramatically resulting in the availability of data on human microbiome composition and function for different areas of problems. Such studies provide the raw data to deeply explore host-microbiome associations and their relation to the development and progression of various complex diseases, environmental phenomena, and abilities or features. Advanced data analysis approaches are needed for the comprehensive utilization and investigation of all data available from these studies, keeping in mind the intrinsic nature of microbiome data when it comes to composition and functional capabilities. Taxonomy-aware approaches and analysis pipelines can potentially help detect the hidden associations and patterns between microbiome and case states, such as for a disease, in order to develop customized prediction models.
The development of prediction models that are customized or tailored to solve a certain problem can be realized via Machine Learning (ML). ML algorithms allow for the development of classification models that can predict a disease state after being trained on a set of informative biomarkers from the microbial communities used to represent each sample of a known group. In order to leverage ML into more translational research related to the microbiome and strengthen our ability to extract meaningful and useful patterns, it is important for the models’ results to be interpretable. The feature space that is built of informative features or biomarkers can be leveraged for this purpose. The investigation should not be limited to only the composition of the space of biomarkers, but also how they act or contribute to the observed phenotypes. This opens up the possibility to incorporate the functional component into the space of features used for training and observation interpretation. Advances in ML algorithms and preprocessing methods, such as feature selection and engineering, allow for analysis of the microbiome that leads to interpretable outcomes and phenotypes, which can be utilized for the development of microbe-targeted solutions, such as therapies and ultimately personalized medicine.
Topics of interest include the following:
1. Knowledge representation and reasoning of microbial data
2. Applications of machine learning in microbiome studies
3. Feature selection and engineering for biomarker identification
4. Microbiome analysis for disease prediction
5. Microbial data analysis for predicting gene functionality
Keywords:
microbiome analysis, gene functionality, machine learning, microbial data
Important Note:
All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.