The sbv IMPROVER Systems Toxicology Computational Challenge: Identification of Human and Species-Independent Blood Response Markers as Predictors of Smoking Exposure and Cessation Status

Blood and urine multi-omics analysis of the impact of e-vaping, smoking, and cessation: from exposome to molecular responses

Cigarette smoking is a major preventable cause of morbidity and mortality. While quitting smoking is the best option, switching from cigarettes to non-combustible alternatives (NCAs) such as e-vapor products is a viable harm reduction approach for smokers who would otherwise continue to smoke. A key challenge for the clinical assessment of NCAs is that self-reported product use can be unreliable, compromising the proper evaluation of their risk reduction potential. In this cross-sectional study of 205 healthy volunteers, we combined comprehensive exposure characterization with in-depth multi-omics profiling to compare effects across four study groups: cigarette smokers (CS), e-vapor users (EV), former smokers (FS), and never smokers (NS). Multi-omics analyses included metabolomics, transcriptomics, DNA methylomics, proteomics, and lipidomics. Comparison of the molecular effects between CS and NS recapitulated several previous observations, such as increased inflammatory markers in CS. Generally, FS and EV demonstrated intermediate molecular effects between the NS and CS groups. Stratification of the FS and EV by combustion exposure markers suggested that this position on the spectrum between CS and NS was partially driven by non-compliance/dual use. Overall, this study highlights the importance of in-depth exposure characterization before biological effect characterization for any NCA assessment study.

Results and lessons learned from the sbv IMPROVER metagenomics diagnostics for inflammatory bowel disease challenge

A growing body of evidence links gut microbiota changes with inflammatory bowel disease (IBD), raising the potential benefit of exploiting metagenomics data for non-invasive IBD diagnostics. The sbv IMPROVER metagenomics diagnosis for inflammatory bowel disease challenge investigated computational metagenomics methods for discriminating IBD and nonIBD subjects. Participants in this challenge were given independent training and test metagenomics data from IBD and nonIBD subjects, which could be wither either raw read data (sub-challenge 1, SC1) or processed Taxonomy- and Function-based profiles (sub-challenge 2, SC2). A total of 81 anonymized submissions were received between September 2019 and March 2020. Most participants' predictions performed better than random predictions in classifying IBD versus nonIBD, Ulcerative Colitis (UC) versus nonIBD, and Crohn's Disease (CD) versus nonIBD. However, discrimination between UC and CD remains challenging, with the classification quality similar to the set of random predictions. We analyzed the class prediction accuracy, the metagenomics features by the teams, and computational methods used. These results will be openly shared with the scientific community to help advance IBD research and illustrate the application of a range of computational methodologies for effective metagenomic classification.

Crowdsourced benchmarking of taxonomic metagenome profilers: lessons learned from the sbv IMPROVER Microbiomics challenge

Background

Selection of optimal computational strategies for analyzing metagenomics data is a decisive step in determining the microbial composition of a sample, and this procedure is complex because of the numerous tools currently available. The aim of this research was to summarize the results of crowdsourced sbv IMPROVER Microbiomics Challenge designed to evaluate the performance of off-the-shelf metagenomics software as well as to investigate the robustness of these results by the extended post-challenge analysis. In total 21 off-the-shelf taxonomic metagenome profiling pipelines were benchmarked for their capacity to identify the microbiome composition at various taxon levels across 104 shotgun metagenomics datasets of bacterial genomes (representative of various microbiome samples) from public databases. Performance was determined by comparing predicted taxonomy profiles with the gold standard.

Results

Most taxonomic profilers performed homogeneously well at the phylum level but generated intermediate and heterogeneous scores at the genus and species levels, respectively. kmer-based pipelines using Kraken with and without Bracken or using CLARK-S performed best overall, but they exhibited lower precision than the two marker-gene-based methods MetaPhlAn and mOTU. Filtering out the 1% least abundance species—which were not reliably predicted—helped increase the performance of most profilers by increasing precision but at the cost of recall. However, the use of adaptive filtering thresholds determined from the sample’s Shannon index increased the performance of most kmer-based profilers while mitigating the tradeoff between precision and recall.

Conclusions

kmer-based metagenomic pipelines using Kraken/Bracken or CLARK-S performed most robustly across a large variety of microbiome datasets. Removing non-reliably predicted low-abundance species by using diversity-dependent adaptive filtering thresholds further enhanced the performance of these tools. This work demonstrates the applicability of computational pipelines for accurately determining taxonomic profiles in clinical and environmental contexts and exemplifies the power of crowdsourcing for unbiased evaluation.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12864-022-08803-2.

The amniotic fluid proteome predicts imminent preterm delivery in asymptomatic women with a short cervix

Preterm birth, the leading cause of perinatal morbidity and mortality, is associated with increased risk of short- and long-term adverse outcomes. For women identified as at risk for preterm birth attributable to a sonographic short cervix, the determination of imminent delivery is crucial for patient management. The current study aimed to identify amniotic fluid (AF) proteins that could predict imminent delivery in asymptomatic patients with a short cervix. This retrospective cohort study included women enrolled between May 2002 and September 2015 who were diagnosed with a sonographic short cervix (< 25 mm) at 16-32 weeks of gestation. Amniocenteses were performed to exclude intra-amniotic infection; none of the women included had clinical signs of infection or labor at the time of amniocentesis. An aptamer-based multiplex platform was used to profile 1310 AF proteins, and the differential protein abundance between women who delivered within two weeks from amniocentesis, and those who did not, was determined. The analysis included adjustment for quantitative cervical length and control of the false-positive rate at 10%. The area under the receiver operating characteristic curve was calculated to determine whether protein abundance in combination with cervical length improved the prediction of imminent preterm delivery as compared to cervical length alone. Of the 1,310 proteins profiled in AF, 17 were differentially abundant in women destined to deliver within two weeks of amniocentesis independently of the cervical length (adjusted p-value < 0.10). The decreased abundance of SNAP25 and the increased abundance of GPI, PTPN11, OLR1, ENO1, GAPDH, CHI3L1, RETN, CSF3, LCN2, CXCL1, CXCL8, PGLYRP1, LDHB, IL6, MMP8, and PRTN3 were associated with an increased risk of imminent delivery (odds ratio > 1.5 for each). The sensitivity at a 10% false-positive rate for the prediction of imminent delivery by a quantitative cervical length alone was 38%, yet it increased to 79% when combined with the abundance of four AF proteins (CXCL8, SNAP25, PTPN11, and MMP8). Neutrophil-mediated immunity, neutrophil activation, granulocyte activation, myeloid leukocyte activation, and myeloid leukocyte-mediated immunity were biological processes impacted by protein dysregulation in women destined to deliver within two weeks of diagnosis. The combination of AF protein abundance and quantitative cervical length improves prediction of the timing of delivery compared to cervical length alone, among women with a sonographic short cervix.

Netpro2vec: A Graph Embedding Framework for Biomedical Applications

The ever-increasing importance of structured data in different applications, especially in the biomedical field, has driven the need for reducing its complexity through projections into a more manageable space. The latest methods for learning features on graphs focus mainly on the neighborhood of nodes and edges. Methods capable of providing a representation that looks beyond the single node neighborhood are kernel graphs. However, they produce handcrafted features unaccustomed with a generalized model. To reduce this gap, in this work we propose a neural embedding framework, based on probability distribution representations of graphs, named Netpro2vec. The goal is to look at basic node descriptions other than the degree, such as those induced by the Transition Matrix and Node Distance Distribution. Netpro2vec provides embeddings completely independent from the task and nature of the data. The framework is evaluated on synthetic and various real biomedical network datasets through a comprehensive experimental classification phase and is compared to well-known competitors.

A comparative study of machine learning methods for predicting the evolution of brain connectivity from a baseline timepoint

BACKGROUND

Predicting the evolution of the brain network, also called connectome, by foreseeing changes in the connectivity weights linking pairs of anatomical regions makes it possible to spot connectivity-related neurological disorders in earlier stages and detect the development of potential connectomic anomalies. Remarkably, such a challenging prediction problem remains least explored in the predictive connectomics literature. It is a known fact that machine learning (ML) methods have proven their predictive abilities in a wide variety of computer vision problems. However, ML techniques specifically tailored for the prediction of brain connectivity evolution trajectory from a single timepoint are almost absent.

NEW METHOD

To fill this gap, we organized a Kaggle competition where 20 competing teams designed advanced machine learning pipelines for predicting the brain connectivity evolution from a single timepoint. The teams developed their ML pipelines with combination of data pre-processing, dimensionality reduction and learning methods. Each ML framework inputs a baseline brain connectivity matrix observed at baseline timepoint t₀ and outputs the brain connectivity map at a follow-up timepoint t₁. The longitudinal OASIS-2 dataset was used for model training and evaluation. Both random data split and 5-fold cross-validation strategies were used for ranking and evaluating the generalizability and scalability of each competing ML pipeline.

RESULTS

Utilizing an inclusive approach, we ranked the methods based on two complementary evaluation metrics (mean absolute error (MAE) and Pearson Correlation Coefficient (PCC)) and their performances using different training and testing data perturbation strategies (single random split and cross-validation). The final rank was calculated using the rank product for each competing team across all evaluation measures and validation strategies. Furthermore, we added statistical significance values to each proposed pipeline.

CONCLUSION

In support of open science, the developed 20 ML pipelines along with the connectomic dataset are made available on GitHub (https://github.com/basiralab/Kaggle-BrainNetPrediction-Toolbox). The outcomes of this competition are anticipated to lead the further development of predictive models that can foresee the evolution of the brain connectivity over time, as well as other types of networks (e.g., genetic networks).

Machine learning methods for brain network classification: Application to autism diagnosis using cortical morphological networks

BACKGROUND

Autism spectrum disorder (ASD) affects the brain connectivity at different levels. Nonetheless, non-invasively distinguishing such effects using magnetic resonance imaging (MRI) remains very challenging to machine learning diagnostic frameworks due to ASD heterogeneity. So far, existing network neuroscience works mainly focused on functional (derived from functional MRI) and structural (derived from diffusion MRI) brain connectivity, which might not directly capture relational morphological changes between brain regions. Indeed, machine learning (ML) studies for ASD diagnosis using morphological brain networks derived from conventional T1-weighted MRI are very scarce.

NEW METHOD

To fill this gap, we leverage crowdsourcing by organizing a Kaggle competition to build a pool of machine learning pipelines for neurological disorder diagnosis with application to ASD diagnosis using cortical morphological networks derived from T1-weighted MRI.

RESULTS

During the competition, participants were provided with a training dataset and only allowed to check their performance on a public test data. The final evaluation was performed on both public and hidden test datasets based on accuracy, sensitivity, and specificity metrics. Teams were ranked using each performance metric separately and the final ranking was determined based on the mean of all rankings. The first-ranked team achieved 70% accuracy, 72.5% sensitivity, and 67.5% specificity, where the second-ranked team achieved 63.8%, 62.5%, 65% respectively.

CONCLUSION

Leveraging participants to design ML diagnostic methods within a competitive machine learning setting has allowed the exploration and benchmarking of wide spectrum of ML methods for ASD diagnosis using cortical morphological networks.

Application of Bayesian networks in determining nanoparticle-induced cellular outcomes using transcriptomics

Inroads have been made in our understanding of the risks posed to human health and the environment by nanoparticles (NPs) but this area requires continuous research and monitoring. Machine learning techniques have been applied to nanotoxicology with very encouraging results. This study deals with bridging physicochemical properties of NPs, experimental exposure conditions and in vitro characteristics with biological effects of NPs on a molecular cellular level from transcriptomics studies. The bridging is done by developing and implementing Bayesian Networks (BNs) with or without data preprocessing. The BN structures are derived either automatically or methodologically and compared. Early stage nanotoxicity measurements represent a challenge, not least when attempting to predict adverse outcomes and modeling is critical to understanding the biological effects of exposure to NPs. The preprocessed data-driven BN showed improved performance over automatically structured BN and the BN with unprocessed datasets. The prestructured BN captures inter relationships between NP properties, exposure condition and in vitro characteristics and links those with cellular effects based on statistic correlation findings. Information gain analysis showed that exposure dose, NP and cell line variables were the most influential attributes in predicting the biological effects. The BN methodology proposed in this study successfully predicts a number of toxicologically relevant cellular disrupted biological processes such as cell cycle and proliferation pathways, cell adhesion and extracellular matrix responses, DNA damage and repair mechanisms etc., with a success rate >80%. The model validation from independent data shows a robust and promising methodology for incorporating transcriptomics outcomes in a hazard and, by extension, risk assessment modeling framework by predicting affected cellular functions from experimental conditions.

A Meta-Analysis of the Performance of a Blood-Based Exposure Response Gene Signature Across Clinical Studies on the Tobacco Heating System 2.2 (THS 2.2)

As part of emerging tobacco harm reduction strategies, modified risk tobacco products (MRTP) are being developed to offer alternatives that have the potential to reduce the individual risk and population harm compared with smoking cigarettes for adult smokers who want to continue using tobacco and nicotine products. MRTPs are defined as any tobacco products that are distributed for use to reduce harm or the risk of tobacco-related disease associated with commercially marketed tobacco products. One such candidate MRTP is the Tobacco Heating System (THS) 2.2, which does not burn tobacco but instead heats it, thus producing significantly reduced levels of harmful and potentially harmful constituents compared with cigarettes. The clinical assessment of candidate MRTPs requires the development of exposure-response markers to distinguish current smokers from either nonsmokers or former smokers with high specificity and sensitivity. Toward this end, a whole blood-derived gene signature was previously developed and reported. Four randomized, controlled, open-label, three-arm parallel group reduced exposure clinical studies have been conducted with subjects randomized to three arms: switching from cigarettes to THS 2.2, continuous use of cigarettes, or smoking abstinence. These clinical studies had an investigational period of 5 days in confinement, which was followed by an 85-day ambulatory period in two studies. Here we tested the previously developed blood-derived signature on the samples derived from those clinical studies. We showed that in all four studies, the signature scores were reduced consistently in subjects who either stopped smoking or switched to THS 2.2 compared with subjects who continued smoking cigarettes.

Ensemble of rankers for efficient gene signature extraction in smoke exposure classification

Background

System toxicology aims at understanding the mechanisms used by biological systems to respond to toxicants. Such understanding can be leveraged to assess the risk of chemicals, drugs, and consumer products in living organisms. In system toxicology, machine learning techniques and methodologies are applied to develop prediction models for classification of toxicant exposure of biological systems. Gene expression data (RNA/DNA microarray) are often used to develop such prediction models.

Results

The outcome of the present work is an experimental methodology to develop prediction models, based on robust gene signatures, for the classification of cigarette smoke exposure and cessation in humans. It is a result of the participation in the recent sbv IMPROVER SysTox Computational Challenge. By merging different gene selection techniques, we obtain robust gene signatures and we investigate prediction capabilities of different off-the-shelf machine learning techniques, such as artificial neural networks, linear models and support vector machines. We also predict six novel genes in our signature, and firmly believe these genes have to be further investigated as biomarkers for tobacco smoking exposure.

Conclusions

The proposed methodology provides gene signatures with top-ranked performances in the prediction of the investigated classification methods, as well as new discoveries in genetic signatures for bio-markers of the smoke exposure of humans.

Electronic supplementary material

The online version of this article (10.1186/s12859-018-2035-3) contains supplementary material, which is available to authorized users.