Discovery of sparse, reliable omic biomarkers with Stabl.

Adoption of high-content omic technologies in clinical studies, coupled with computational methods, has yielded an abundance of candidate biomarkers. However, translating such findings into bona fide clinical biomarkers remains challenging. To facilitate this process, we introduce Stabl, a general machine learning method that identifies a sparse, reliable set of biomarkers by integrating noise injection and a data-driven signal-to-noise threshold into multivariable predictive modeling. Evaluation of Stabl on synthetic datasets and five independent clinical studies demonstrates improved biomarker sparsity and reliability compared to commonly used sparsity-promoting regularization methods while maintaining predictive performance; it distills datasets containing 1,400-35,000 features down to 4-34 candidate biomarkers. Stabl extends to multi-omic integration tasks, enabling biological interpretation of complex predictive models, as it hones in on a shortlist of proteomic, metabolomic and cytometric events predicting labor onset, microbial biomarkers of pre-term birth and a pre-operative immune signature of post-surgical infections. Stabl is available at https://github.com/gregbellan/Stabl .

Microbiome preterm birth DREAM challenge: Crowdsourcing machine learning approaches to advance preterm birth research.

Every year, 11% of infants are born preterm with significant health consequences, with the vaginal microbiome a risk factor for preterm birth. We crowdsource models to predict (1) preterm birth (PTB; <37 weeks) or (2) early preterm birth (ePTB; <32 weeks) from 9 vaginal microbiome studies representing 3,578 samples from 1,268 pregnant individuals, aggregated from public raw data via phylogenetic harmonization. The predictive models are validated on two independent unpublished datasets representing 331 samples from 148 pregnant individuals. The top-performing models (among 148 and 121 submissions from 318 teams) achieve area under the receiver operator characteristic (AUROC) curve scores of 0.69 and 0.87 predicting PTB and ePTB, respectively. Alpha diversity, VALENCIA community state types, and composition are important features in the top-performing models, most of which are tree-based methods. This work is a model for translation of microbiome data into clinically relevant predictive models and to better understand preterm birth.

MaLiAmPi enables generalizable and taxonomy-independent microbiome features from technically diverse 16S-based microbiome studies.

For studies using microbiome data, the ability to robustly combine data from technically and biologically distinct microbiome studies is a crucial means of supporting more robust and clinically relevant inferences. Formidable technical challenges arise when attempting to combine data from technically diverse 16S rRNA gene variable region amplicon sequencing (16S) studies. Closed operational taxonomic units and taxonomy are criticized as being heavily dependent upon reference sets and with limited precision relative to the underlying biology. Phylogenetic placement has been demonstrated to be a promising taxonomy-free manner of harmonizing microbiome data, but it has lacked a validated count-based feature suitable for use in machine learning and association studies. Here we introduce a phylogenetic-placement-based, taxonomy-independent, compositional feature of microbiota: phylotypes. Phylotypes were predictive of clinical outcomes such as obesity or pre-term birth on technically diverse independent validation sets harmonized post hoc. Thus, phylotypes enable the rigorous cross-validation of 16S-based clinical prognostic models and associative microbiome studies.

Microbiome Preterm Birth DREAM Challenge: Crowdsourcing Machine Learning Approaches to Advance Preterm Birth Research.

Globally, every year about 11% of infants are born preterm, defined as a birth prior to 37 weeks of gestation, with significant and lingering health consequences. Multiple studies have related the vaginal microbiome to preterm birth. We present a crowdsourcing approach to predict: (a) preterm or (b) early preterm birth from 9 publicly available vaginal microbiome studies representing 3,578 samples from 1,268 pregnant individuals, aggregated from raw sequences via an open-source tool, MaLiAmPi. We validated the crowdsourced models on novel datasets representing 331 samples from 148 pregnant individuals. From 318 DREAM challenge participants we received 148 and 121 submissions for our two separate prediction sub-challenges with top-ranking submissions achieving bootstrapped AUROC scores of 0.69 and 0.87, respectively. Alpha diversity, VALENCIA community state types, and composition (via phylotype relative abundance) were important features in the top performing models, most of which were tree based methods. This work serves as the foundation for subsequent efforts to translate predictive tests into clinical practice, and to better understand and prevent preterm birth.

VMAP: Vaginal Microbiome Atlas During Pregnancy.

The vaginal microbiome has been shown to be associated with pregnancy outcomes including preterm birth (PTB) risk. Here we present VMAP: Vaginal Microbiome Atlas during Pregnancy (http://vmapapp.org), an application to visualize features of 3,909 vaginal microbiome samples of 1,416 pregnant individuals from 11 studies, aggregated from raw public and newly generated sequences via an open-source tool, MaLiAmPi. Our visualization tool (http://vmapapp.org) includes microbial features such as various measures of diversity, VALENCIA community state types (CST), and composition (via phylotypes and taxonomy). This work serves as a resource for the research community to further analyze and visualize vaginal microbiome data in order to better understand both healthy term pregnancies and those associated with adverse outcomes.

Associations between forensic loci and expression levels of neighboring genes may compromise medical privacy.

A set of 20 short tandem repeats (STRs) is used by the US criminal justice system to identify suspects and to maintain a database of genetic profiles for individuals who have been previously convicted or arrested. Some of these STRs were identified in the 1990s, with a preference for markers in putative gene deserts to avoid forensic profiles revealing protected medical information. We revisit that assumption, investigating whether forensic genetic profiles reveal information about gene-expression variation or potential medical information. We find six significant correlations (false discovery rate = 0.23) between the forensic STRs and the expression levels of neighboring genes in lymphoblastoid cell lines. We explore possible mechanisms for these associations, showing evidence compatible with forensic STRs causing expression variation or being in linkage disequilibrium with a causal locus in three cases and weaker or potentially spurious associations in the other three cases. Together, these results suggest that forensic genetic loci may reveal expression levels and, perhaps, medical information.