Biology subjects

Carrieri, A. P.

Publications and source records attributed to Carrieri, A. P..

8 recordsLinked to original sources

bioRxiv2026

RNA foundation models enable generalizable endometriosis disease classification and stable gene-level interpretation

Endometriosis is a chronic inflammatory condition with significant diagnostic delays impacting one in ten reproductive age women worldwide. While machine learning (ML) models trained on transcriptomic data show promise for disease prediction, limited generalizability across independent patient cohorts has hindered clinical translation. Foundations models (FMs) pretrained on large-scale transcriptomic data offer promise to learn transferrable, biologically meaningful representations that could support cross-cohort predictions. We assembled a 12-cohort bulk RNA-seq benchmark (334 samples) and developed a computationally efficient pipeline to test whether FMs improve endometriosis classification, an approach not previously applied to this disease. Using AutoXAI4Omics with cohort-aware validation, we compared embeddings derived from five state-of-the-art RNA FMs against TPM baselines. In cross-cohort prediction, FM embeddings significantly improved performance, achieving a weighted F1-score of 0.83 vs. 0.68 for the baseline. To allow gene-level interpretation of FM embedding models, we introduce classified-aligned integrated gradients (CA-IG), an interpretability approach aligning gene-level attributions to the downstream classifier without end-to-end finetuning. CA-IG revealed a conserved set of predictive genes from FM embeddings across cohort-validation regimes, contrasting with unstable baseline explainability, suggesting that FM embeddings prioritized transferable disease-related signal over cohort-specific effects. These genes include novel candidates that converge on biologically plausible pathways for endometriosis.

bioinformatics↗

bioRxiv2024

AutoXAI4Omics: an Automated Explainable AI tool for Omics and tabular data

Machine learning (ML) methods have the potential of detailed insights of complex biological systems and today are increasingly used to analyse omics data for tasks such as the discovery of novel biomarkers and phenotype prediction. It can be extremely beneficial and powerful for scientists, domain experts, to easily run sophisticated, robust, and interpretable ML pipelines without the need for an in depth understanding of the code needed to train, tune, optimise ML algorithms. They can then focus on the biological interpretation and validation of the results and insights generated by ML models. Here, we present an entirely automated open-source explainable AI tool, AutoXAI4Omics, that performs classification and regression tasks from omics and tabular numerical data. AutoXAI4Omics accelerates scientific discovery by automating processes and decisions made by AI experts, e.g., selection of the best feature set, hyper-tuning of different ML algorithms and selection of the best ML model for a specific task and dataset. Prior to ML analysis AutoXAI4Omics incorporates feature filtering options that are tailored to specific omic data types. Moreover, the insights into the predictions that are provided by the tool through explainability analysis highlight associations between omic feature values and the targets under investigation e.g., predicted phenotypes, facilitating the discovery of actionable insights. AutoXAI4Omics is at: https://github.com/IBM/AutoXAI4Omics. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=188 HEIGHT=200 SRC="FIGDIR/small/586460v1_ufig1.gif" ALT="Figure 1"> View larger version (36K): org.highwire.dtl.DTLVardef@366327org.highwire.dtl.DTLVardef@a7b559org.highwire.dtl.DTLVardef@7319aforg.highwire.dtl.DTLVardef@9b5030_HPS_FORMAT_FIGEXP M_FIG C_FIG

bioinformatics↗

bioRxiv2022

Bacterial metatranscriptomes in wastewater can differentiate virally infected human populations

Monitoring wastewater samples at building-level resolution screens large populations for SARS-CoV-2, prioritizing testing and isolation efforts. Here we perform untargeted metatranscriptomics on virally-enriched wastewater samples from 10 locations on the UC San Diego campus, demonstrating that resulting bacterial taxonomic and functional profiles discriminate SARS-CoV-2 status even without direct detection of viral transcripts. Our proof-of-principle reveals emergent threats through changes in the human microbiome, suggesting new approaches for untargeted wastewater-based epidemiology.

microbiology↗

bioRxiv2021

Multi-omics profiling of Earth's biomes reveals that microbial and metabolite composition are shaped by the environment

As our understanding of the structure and diversity of the microbial world grows, interpreting its function is of critical interest for understanding and managing the many systems microbes influence. Despite advances in sequencing, lack of standardization challenges comparisons among studies that could provide insight into the structure and function of microbial communities across multiple habitats on a planetary scale. Technical variation among distinct studies without proper standardization of approaches prevents robust meta-analysis. Here, we present a multi-omics, meta-analysis of a novel, diverse set of microbial community samples collected for the Earth Microbiome Project. We include amplicon (16S, 18S, ITS) and shotgun metagenomic sequence data, and untargeted metabolomics data (liquid chromatography-tandem mass spectrometry and gas chromatography mass spectrometry), centering our description on relationships and co-occurrences of microbially-related metabolites and microbial taxa across environments. Standardized protocols and analytical methods for characterizing microbial communities, including assessment of molecular diversity using untargeted metabolomics, facilitate identification of shared microbial and metabolite features, permitting us to explore diversity at extraordinary scale. In addition to a reference database for metagenomic and metabolomic data, we provide a framework for incorporating additional studies, enabling the expansion of existing knowledge in the form of a community resource that will become more valuable with time. To provide examples of applying this database, we outline important ecological questions that can be addressed, and test the hypotheses that every microbe and metabolite is everywhere, but the environment selects. Our results show that metabolite diversity exhibits turnover and nestedness related to both microbial communities and the environment. The relative abundances of microbially-related metabolites vary and co-occur with specific microbial consortia in a habitat-specific manner, and highlight the power of certain chemistry - in particular terpenoids - in distinguishing Earths environments.

ecology↗

bioRxiv2021

Interpreting machine learning models to investigate circadian regulation and facilitate exploration of clock function

The circadian clock is an important adaptation to life on earth. Here, we use machine learning to predict complex temporal circadian gene expression patterns in Arabidopsis. Most significantly, we classify circadian genes using DNA sequence features generated from public genomic resources, with no experimental work or prior knowledge needed. We use model explanation to rank DNA sequence features, observing transcript-specific combinations of potential circadian regulatory elements that discriminate temporal phase of expression. Model interpretation/explanation provides the backbone of our methodological advances, giving insight into biological processes and experimental design. Next, we use model interpretation to optimize sampling strategies when we predict circadian transcripts using reduced numbers of transcriptomic timepoints, saving both time and money. Finally, we predict the circadian time from a single transcriptomic timepoint, deriving novel marker transcripts that are most impactful for accurate prediction, this could facilitate the identification of altered clock function from existing datasets.

genomics↗

bioRxiv2020

EMPress enables tree-guided, interactive, and exploratory analyses of multi-omic datasets

Standard workflows for analyzing microbiomes often include the creation and curation of phylogenetic trees. Here we present EMPress, an interactive tool for visualizing trees in the context of microbiome, metabolome, etc. community data scalable beyond modern large datasets like the Earth Microbiome Project. EMPress provides novel functionality--including ordination integration and animations--alongside many standard tree visualization features, and thus simplifies exploratory analyses of many forms of omic data.

bioinformatics↗

bioRxiv2020

Explainable AI reveals key changes in skin microbiome associated with menopause, smoking, aging and skin hydration

Alterations in the human microbiome have been observed in a variety of conditions such has asthma, gingivitis, dermatitis and cancer, and much remains to be learned about the links between the microbiome and human health. The fusion of artificial intelligence with rich microbiome datasets can offer an improved understanding of the microbiomes role in our health. To gain actionable insights it is essential to consider both the predictive power and the transparency of the models by providing explanations for the predictions. We combine the effort of collecting a corpus of leg skin microbiome samples of two healthy cohorts of women with the development of an explainable artificial intelligence (EAI) approach that provides accurate predictions of phenotypes and explanations. The explanations are expressed in terms of variations in the abundance of key microbes that drive the predictions. We predict skin hydration, subjects age, pre/post-menopausal status and smoking status from the leg skin microbiome. The key changes in microbial composition linked to skin hydration can accelerate the development of personalised treatments for healthy skin, while those associated with age may offer insights into the skin aging process. The leg microbiome signatures associated with smoking and menopausal status are consistent with previous findings from oral/respiratory tract microbiomes and vaginal microbiomes respectively. This suggests that easily accessible microbiome samples could be used to investigate health-related phenotypes, offering potential for non-invasive diagnosis and condition monitoring. Our EAI approach sets the stage for new work focused on understanding the complex relationships between microbial communities and phenotypes. Our approach can be applied to predict any conditions from microbiome samples and has the potential to accelerate the development of microbiome-based personalised therapeutics and non-invasive diagnostics.

bioinformatics↗

bioRxiv2018

Streaming histogram sketching for rapid microbiome analytics

MotivationThe growth in publically available microbiome data in recent years has yielded an invaluable resource for genomic research; allowing for the design of new studies, augmentation of novel datasets and reanalysis of published works. This vast amount of microbiome data, as well as the widespread proliferation of microbiome research and the looming era of clinical metagenomics, means there is an urgent need to develop analytics that can process huge amounts of data in a short amount of time.\n\nTo address this need, we propose a new method for the compact representation of microbiome sequencing data using similarity-preserving sketches of streaming k-mer spectra. These sketches allow for dissimilarity estimation, rapid microbiome catalogue searching, and classification of microbiome samples in near real-time.\n\nResultsWe apply streaming histogram sketching to microbiome samples as a form of dimensionality reduction, creating a compressed histosketch that can be used to efficiently represent microbiome k-mer spectra. Using public microbiome datasets, we show that histosketches can be clustered by sample type using pairwise Jaccard similarity estimation, consequently allowing for rapid microbiome similarity searches via a locality sensitive hashing indexing scheme. Furthermore, we show that histosketches can be used to train machine learning classifiers to accurately label microbiome samples. Specifically, using a collection of 108 novel microbiome samples from a cohort of premature neonates, we trained and tested a Random Forest Classifier that could accurately predict whether the neonate had received antibiotic treatment (95% accuracy, precision 97%) and could subsequently be used to classify microbiome data streams in less than 12 seconds.\n\nWe provide our implementation, Histosketching Using Little K-mers (HULK), which can histosketch a typical 2GB microbiome in 50 seconds on a standard laptop using 4 cores, with the sketch occupying 3000 bytes of disk space.\n\nAvailabilityOur implementation (HULK) is written in Go and is available at: https://github.com/will-rowe/hulk (MIT License)

genomics↗