Infant Microbiome Modeling
- We have a set of
control variablesnamely connected to the mode of feeding (3 modes) - We have a set of
internal stateobservations in the form of bacterial colonization profiles - We have one or more
outcome measuressuch as head circumeference measurement as a proxy of cognitive development - All these measurements have tagged timestamps
Aim one: Reverse-engineering the Wiring between control to state to oucome
Aim two: How Does Perturbations in Internal States or Control Variables Affect the Rest of the Parameters
How do we incorporate time in the analysis
Can we carry out modeling at the highest entity resolution?
Relative Biome Dynamics
@article{lee2020exploring,
title={Exploring the Role of Gut Bacteria in Health and Disease in Preterm Neonates},
author={Lee, Jimmy Kok-Foo and Hern Tan, Loh Teng and Ramadas, Amutha and Ab Mutalib, Nurul-Syakima and Lee, Learn-Han},
journal={International journal of environmental research and public health},
volume={17},
number={19},
pages={6963},
year={2020},
publisher={Multidisciplinary Digital Publishing Institute}
}...the postnatal very preterm gut is colonised in a structured sequence by facultative anaerobes followed by obligate anaerobes. The initial class is Bacilli followed by a bloom of Gammaproteobacteria and finally an increase in Clostridia. There is a paucity of Lactobacillus,Bifidobacillus and the phyla Bacteroidetes and Actinobacteria.
- The final pattern is well established by 30 to 36 weeks corrected age.
- At phylum level, Firmicutes dominates at 1 week and Proteobacteria at 1 month.
- The prevalence of Firmicutes at 1 week is 0.65 decreasing to 0.32 by 1 month while Proteobacteria increases from 0.28 to 0.66 over the same period.
The Shannon diversity index ranges from 0.38 to 1.4 at 1 week, increases by 2 weeks and reaches 0.71 to 0.9 by 1 month.
- measure the Shannon diversity index
@article{pammi2017intestinal,
title={Intestinal dysbiosis in preterm infants preceding necrotizing enterocolitis: a systematic review and meta-analysis},
author={Pammi, Mohan and Cope, Julia and Tarr, Phillip I and Warner, Barbara B and Morrow, Ardythe L and Mai, Volker and Gregory, Katherine E and Kroll, J Simon and McMurtry, Valerie and Ferris, Michael J and others},
journal={Microbiome},
volume={5},
number={1},
pages={31},
year={2017},
publisher={Springer}
}
File here
Microbial dysbiosis preceding NEC in preterm infants is characterized by increased relative abundances of Proteobacteria and decreased relative abundances of Firmicutes and Bacteroidetes. Microbiome optimization may provide a novel strategy for preventing NEC.
causal analysis and directional inference
@article{sazal2021causal,
title={Causal effects in microbiomes using interventional calculus},
author={Sazal, Musfiqur and Stebliankin, Vitalii and Mathee, Kalai and Yoo, Changwon and Narasimhan, Giri},
journal={Scientific reports},
volume={11},
number={1},
pages={1--15},
year={2021},
publisher={Nature Publishing Group}
}
@article{john2016gut,
title={The gut microbiome and obesity},
author={John, George Kunnackal and Mullin, Gerard E},
journal={Current oncology reports},
volume={18},
number={7},
pages={1--7},
year={2016},
publisher={Springer}
}
@article{li2017gut,
title={The gut microbiota and autism spectrum disorders},
author={Li, Qinrui and Han, Ying and Dy, Angel Belle C and Hagerman, Randi J},
journal={Frontiers in cellular neuroscience},
pages={120},
year={2017},
publisher={Frontiers}
}
@article{honda2012microbiome,
title={The microbiome in infectious disease and inflammation},
author={Honda, Kenya and Littman, Dan R},
journal={Annual review of immunology},
volume={30},
pages={759},
year={2012},
publisher={NIH Public Access}
}
@article{aarts2017gut,
title={Gut microbiome in ADHD and its relation to neural reward anticipation},
author={Aarts, Esther and Ederveen, Thomas HA and Naaijen, Jilly and Zwiers, Marcel P and Boekhorst, Jos and Timmerman, Harro M and Smeekens, Sanne P and Netea, Mihai G and Buitelaar, Jan K and Franke, Barbara and others},
journal={PloS one},
volume={12},
number={9},
pages={e0183509},
year={2017},
publisher={Public Library of Science San Francisco, CA USA}
}
@article{fischbach2018microbiome,
title={Microbiome: focus on causation and mechanism},
author={Fischbach, Michael A},
journal={Cell},
volume={174},
number={4},
pages={785--790},
year={2018},
publisher={Elsevier}
}
@inproceedings{sazal2018inferring,
title={Inferring relationships in microbiomes from signed bayesian networks},
author={Sazal, Musfiqur Rahman and Ruiz-Perez, Daniel and Cickovski, Trevor and Narasimhan, Giri},
booktitle={2018 IEEE 8th International Conference on Computational Advances in Bio and Medical Sciences (ICCABS)},
pages={1--1},
year={2018},
organization={IEEE}
}
@article{faust2015metagenomics,
title={Metagenomics meets time series analysis: unraveling microbial community dynamics},
author={Faust, Karoline and Lahti, Leo and Gonze, Didier and De Vos, Willem M and Raes, Jeroen},
journal={Current opinion in microbiology},
volume={25},
pages={56--66},
year={2015},
publisher={Elsevier}
}
@article{matchado2021network,
title={Network analysis methods for studying microbial communities: A mini review},
author={Matchado, Monica Steffi and Lauber, Michael and Reitmeier, Sandra and Kacprowski, Tim and Baumbach, Jan and Haller, Dirk and List, Markus},
journal={Computational and structural biotechnology journal},
volume={19},
pages={2687--2698},
year={2021},
publisher={Elsevier}
}
https://www.nature.com/articles/s41598-021-84905-3
With advancements in high-throughput sequencing technology, it is now possible to examine microbial diversity in microbiomes with increased precision, and has elucidated associations between the microbiome and phenotypes such as obesity, neurological disorders, inflammation, immune disorders, metabolic diseases, and more~\cite{john2016gut,li2017gut,honda2012microbiome,aarts2017gut}. Interest in constructing causal networks for microbiomes is recent, and focused experiments in the laboratory to elicit causal relationships within microbiomes do exist~\cite{fischbach2018microbiome}, but do not employ computational causal inferencing approaches. Sazal et al. were among the first to construct causal networks for microbiomes~\cite{sazal2018inferring}.
Network analysis methods for studying microbial communities: A mini review ~\cite{matchado2021network} Threee important biases: compositionality, sparsity and spurious correlations in microbial co-occurrence network analysis. Firstly, microbiome data are compositional [11]; i.e. microbial counts represent proportions instead of absolute abundances. Secondly, sparsity in the dataset can lead to false associations of microorganisms. A zero indicates either the absence of a microorganism, or an insufficient sequencing depth. Thirdly, it is challenging to differentiate between direct and indirect associations, in particular if these are related to environmental factors.
Correlation-based techniques, including Pearson or Spearman correlation, are among the most popular methods for studying microbial interactions in human gut [12], oral [13] and soil [14] microbiomes. Weiss et al. [15] evaluated the strengths and weaknesses of eight different correlation methods and provided recommendations based on the nature of the data and identified sparsity as a key issue not sufficiently addressed by these approaches. Correlation analysis often results in artefactual and spurious associations between low-abundant microbial members in a community as it fails to account for compositionality [11]. As Lovell et al. [16], [17] showed, correlation-based methods are not subcompositionally coherent such that, for instance, depleting rare taxa is expected to change the outcome of correlation analysis. To overcome this issue, compositional data analysis can be employed. Various proportionality measures [16], [17] have been proposed some of which are implemented in the R package propr [18] and can be used for network construction. A frequently used method to account for compositionality is centered log ratio transformation (CLR) [19], [20], where the geometric mean of the sample vector is used as the reference. CLR transformation maps the relative counts from simplex into Euclidean space and hence makes these data compatible with linear analysis methods. Apart from these classical approaches, more complex methods have been proposed based on probabilistic graphical, Gaussian graphical and complex multiple regression models to construct microbial interaction networks [6], [21]. Most methods take compositionality into account either by performing CLR transformation as a pre-processing step or by using a Dirichlet multinomial model to directly account for compositionality. Existing methods differ with respect to sensitivity, specificity and computational complexity and can be grouped into four different categories (Fig. 2).
| algorithm | limitation |
|---|---|
| MENAP (2012) | does not address compositionaity bias, sparsity |
| SparCC (2012) | Nonlinear relationships cannot be detected |
| CCLasso (2015) | Nonlinear relationships cannot be detected |
| REBACCA (2015) | linear |
| CoNet (2016) | does not address compositionaity bias |
| Meta-network (2019) | does not address compositionaity bias |
| Correlation-Centric Network (2020) | does not address compositionaity bias |
| MDiNE (2019) | single binary covariate to construct the networks |
** correlation-based methods** Many correlation-based methods employ variants of Pearson or Spearman correlation to obtain an estimate of microbial interaction between pairs of taxa [24], [25]. However, these measures do not account for compositionality, where, for instance, an increase in absolute abundance of just a single taxon is followed by a decrease in relative abundances of all other taxa even if their absolute abundance does not change. This can be mitigated by ratio transformation of the data. Ratio transformations ensure that the ratios between two features are the same whether the data are absolute counts or proportions. Taking the logarithm of these counts makes the data further symmetric and linearly related [19]. The resulting correlation coefficients are thus compositionally coherent, i.e. the log ratio of two taxa is completely independent of other taxa.
** regularized linera regression** An alternative to correlation methods is to build linear regression models in which the abundance of each taxon is modelled as a response variable using the abundance of all other taxa as explanatory variables. Here, the coefficient of each taxon serves as a linear measure for the interaction strength of two taxa. However, due to the large number of features, such models are generally prone to overfitting.
** conditional dependence and graphical methods ** Partial correlation [40] and related approaches are used here to distinguish between direct and indirect interactions, resulting in an undirected weighted graph where the edges imply the conditional dependency between two taxa.
** time series analysis for microbiomes **
~\cite{faust2015metagenomics} Although many standard approaches for longitudinal analysis require long time series with short and regular sampling intervals, the currently available metagenomic time series tend to be short (few time points), gapped (missing time points), sparse (zero-rich) and noisy, necessitating preprocessing steps such as filtering, standardizing, interpolation and detrending to make time points equidistant and comparable.