bioRxiv Science⌕ Search

Biology subjects

Rosen, G. L.

Publications and source records attributed to Rosen, G. L..

11 recordsLinked to original sources
bioRxiv2026

Improving genomic language model reliability under distribution shift

Transformer-based Genomic Language Models (GLMs) have achieved strong performance across diverse genomic prediction tasks. However, their tendency toward overconfident predictions--particularly on noisy or unfamiliar data--limits reliability. In genomics, where unknown species and novel variants are common, developing models robust to distribution shift is crucial for dependable predictions. Here, we analyze the impact of several common and novel uncertainty quantification (UQ) methods in the context of GLMs, evaluating their performance across diverse downstream genomic and metagenomic prediction tasks. Comparing model behavior on both in-distribution (ID) and out-of-distribution (OOD) data, we show that temperature scaling and epistemic neural networks are capable of improving classification reliability across multiple GLM architectures and domains. The software is available at: https://github.com/EESI/glm-epinet-pyt

bioinformatics↗
bioRxiv2024

iSeqsSearch: Incremental Protein Search for iBlast/iMMSeqs2/iDiamond

BackgroundThe advancement of sequencing technology has led to a rapid increase in the amount of DNA and protein sequence data; consequently, the size of genomic and proteomic databases is constantly growing. As a result, database searches need to be continually updated to account for the new data being added. Continually re-searching the entire existing dataset, however, wastes resources. Incremental database search can address this problem. MethodsOne recently introduced incremental search method is iBlast, which wraps the BLAST sequence search method with an algorithm to reuse previously processed data and thereby increase search efficiency. The iBlast wrapper, however, must be generalized to support more performant DNA/protein sequence search methods that have been developed, namely MMseqs2 and Diamond. Moreover, the previously published iBlast wrapper has to be revised to be more robust and usable by the general community. ResultsiMMseqs2 and iDiamond, which apply the incremental approach, obtain results nearly identical to those achieved using only MMseqs2 and Diamond. Notably, when comparing ranking comparison methods such as the Pearson correlation, we observe a high concordance of over 0.9, indicating similar results. Moreover, in some cases, our incremental approach applying iBlast merge function and using m8 formats including the new m8e format provides more hits compared to the conventional MMseqs2 and Diamond. ConclusionThe incremental approach using iMMseqs2 and iDiamond demonstrates efficiency in terms of reusing previously processed data while maintaining high accuracy and concordance in search results. This method can reduce resource waste in continually growing genomic and proteomic database searches. The sample codes are made available at GitHub: https://github.com/EESI/Incremental-Protein-Search.

bioinformatics↗
bioRxiv2024

The Naive Bayes Classifier++ for Metagenomic Taxonomic Classification -- Query Evaluation

This study examines the query performance of the NBC++ (Incremental Naive Bayes Classifier) program for variations in canonicality, kmer size, databases, and input sample data size. NBC++ can successfully assess a wide range of superkingdoms using a small training database. We demonstrate that NBC++ and Kraken2 are affected by database depth with macro measures increasing with depth but that the full diversity of life, especially viruses, is still a challenge for these classifiers. NBC++ spends less time training but at the cost of long querying time. The major enhancements are to accommodate canonical kmer storage (with major storage savings), adaptable and optimized memory allocation that quickens the query analysis and allows the classifier to be run on almost any system, and enables output of the log-likelihood values against each training genome which provides users with valualbe confidence information.

bioinformatics↗
bioRxiv2021

Critical Assessment of Metagenome Interpretation - the second round of challenges

Evaluating metagenomic software is key for optimizing metagenome interpretation and focus of the community-driven initiative for the Critical Assessment of Metagenome Interpretation (CAMI). In its second challenge, CAMI engaged the community to assess their methods on realistic and complex metagenomic datasets with long and short reads, created from [~]1,700 novel and known microbial genomes, as well as [~]600 novel plasmids and viruses. Altogether 5,002 results by 76 program versions were analyzed, representing a 22x increase in results. Substantial improvements were seen in metagenome assembly, some due to using long-read data. The presence of related strains still was challenging for assembly and genome binning, as was assembly quality for the latter. Taxon profilers demonstrated a marked maturation, with taxon profilers and binners excelling at higher bacterial taxonomic ranks, but underperforming for viruses and archaea. Assessment of clinical pathogen detection techniques revealed a need to improve reproducibility. Analysis of program runtimes and memory usage identified highly efficient programs, including some top performers with other metrics. The CAMI II results identify current challenges, but also guide researchers in selecting methods for specific analyses.

bioinformatics↗
bioRxiv2019

Keeping up with the genomes: efficient learning of our increasing knowledge of the tree of life

It is a computational challenge for current metagenomic classifiers to keep up with the pace of training data generated from genome sequencing projects, such as the exponentially-growing NCBI RefSeq bacterial genome database. When new reference sequences are added to training data, statically trained classifiers must be rerun on all data, resulting in a highly inefficient process. The rich literature of "incremental learning" addresses the need to update an existing classifier to accommodate new data without sacrificing much accuracy compared to retraining the classifier with all data. We demonstrate how classification improves over time by incrementally training a classifier on progressive RefSeq snapshots and testing it on: (a) all known current genomes (as a ground truth set) and (b) a real experimental metagenomic gut sample. We demonstrate that as a classifier models knowledge of genomes grows, classification accuracy increases. The proof-of-concept naive Bayes implementation, when updated yearly, now runs in 1/4th of the non-incremental time with no accuracy loss. In conclusion, it is evident that classification improves by having the most current knowledge at its disposal. Therefore, it is of utmost importance to make classifiers computationally tractable to keep up with the data deluge.

bioinformatics↗
bioRxiv2019

Themetagenomics: Exploring Thematic Structure and Predicted Functionality of 16s rRNA Amplicon Data

Analysis of microbiome data involves identifying co-occurring groups of taxa associated with sample features of interest (e.g., disease state). Elucidating such relations is often difficult as microbiome data are compositional, sparse, and have high dimensionality. Also, the configuration of co-occurring taxa may represent overlapping subcommunities that contribute to sample characteristics such as host status. Preserving the configuration of co-occurring microbes rather than detecting specific indicator species is more likely to facilitate biologically meaningful interpretations. Additionally, analyses that use taxonomic relative abundances to predict the abundances of different gene functions aggregate predicted functional profiles across taxa. This precludes straightforward identification of predicted functional components associated with subsets of co-occurring taxa. We provide an approach to explore co-occurring taxa using \"topics\" generated via a topic model and link these topics to specific sample features (e.g., disease state). Rather than inferring predicted functional content based on overall taxonomic relative abundances, we instead focus on inference of functional content within topics, which we parse by estimating interactions between topics and pathways through a multilevel, fully Bayesian regression model. We apply our methods to three publicly available 16S amplicon sequencing datasets: an inflammatory bowel disease dataset from Gevers et al., an oral cancer dataset from Schmidt et al., and a time-series dataset from David et al. Using our topic model approach to uncover latent structure in 16S rRNA amplicon surveys, investigators can (1) capture groups of co-occurring taxa termed topics; (2) uncover within-topic functional potential; (3) link taxa co-occurrence, gene function, and environmental/host features; and (4) explore the way in which sets of co-occurring taxa behave and evolve over time. These methods have been implemented in a freely available R package: https://github.com/EESI/themetagenomics.

bioinformatics↗
bioRxiv2018

16S rRNA sequence embeddings: Meaningful numeric feature representations of nucleotide sequences that are convenient for downstream analyses

Advances in high-throughput sequencing have increased the availability of microbiome sequencing data that can be exploited to characterize microbiome community structure in situ. We explore using word and sentence embedding approaches for nucleotide sequences since they may be a suitable numerical representation for downstream machine learning applications (especially deep learning). This work involves first encoding (\"embedding\") each sequence into a dense, low-dimensional, numeric vector space. Here, we use Skip-Gram word2vec to embed k-mers, obtained from 16S rRNA amplicon surveys, and then leverage an existing sentence embedding technique to embed all sequences belonging to specific body sites or samples. We demonstrate that these representations are biologically meaningful, and hence the embedding space can be exploited as a form of feature extraction for exploratory analysis. We show that sequence embeddings preserve relevant information about the sequencing data such as k-mer context, sequence taxonomy, and sample class. Specifically, the sequence embedding space resolved differences among phyla, as well as differences among genera within the same family. Distances between sequence embeddings had similar qualities to distances between alignment identities, and embedding multiple sequences can be thought of as generating a consensus sequence. Using sample embeddings for body site classification resulted in negligible performance loss compared to using OTU abundance data. Lastly, the k-mer embedding space captured distinct k-mer profiles that mapped to specific regions of the 16S rRNA gene and corresponded with particular body sites. Together, our results show that embedding sequences results in meaningful representations that can be used for exploratory analyses or for downstream machine learning applications that require numeric data. Moreover, because the embeddings are trained in an unsupervised manner, unlabeled data can be embedded and used to bolster supervised machine learning tasks.\n\nAuthor summaryImprovements in the way genomes are sequenced have led to an abundance of microbiome data. With the right approaches, researchers use this data to thoroughly characterize how microbes interact with each other and their host, but sequencing data is of a form (sequences of letters) not ideal for many data analysis approaches. We therefore present an approach to transform sequencing data into arrays of numbers that can capture interesting qualities of the data at the sub-sequence, full-sequence, and sample levels. This allows us to measure the importance of certain microbial sequences with respect to the type of microbe and the condition of the host. Also, representing sequences in this way improves our ability to use other complicated modeling approaches. Using microbiome data from human samples, we show that our numeric representations captured differences between different types of microbes, as well as differences in the body site location from which the samples were collected.

bioinformatics↗
bioRxiv2017

Evaluating a topic model approach for parsing microbiome data structure

The increasing availability of microbiome survey data has led to the use of complex machine learning and statistical approaches to measure taxonomic diversity and extract relationships between taxa and their host or environment. However, many approaches inadequately account for the difficulties inherent to microbiome data. These difficulties include (1) insufficient sequencing depth resulting in sparse count data, (2) a large feature space relative to sample space, resulting in data prone to overfitting, (3) library size imbalance, requiring normalization strategies that lead to compositional artifacts, and (4) zero-inflation. Recent work has used probabilistic topics models to more appropriately model microbiome data, but a thorough inspection of just how well topic models capture underlying microbiome signal is lacking. Also, no work has determined whether library size or variance normalization improves model fitting. Here, we assessed a topic model approach on 16S rRNA gene survey data. Through simulation, we show, for small sample sizes, library-size or variance normalization is unnecessary prior to fitting the topic model. In addition, by exploiting topic-to-topic correlations, the topic model successfully captured dynamic time-series behavior of simulated taxonomic subcommunities. Lastly, when the topic model was applied to the David et al. time-series dataset, three distinct gut configurations emerged. However, unlike the David et al. approach, we characterized the events in terms of topics, which captured taxonomic co-occurrence, and posterior uncertainty, which facilitated the interpretation of how the taxonomic configurations evolved over time.

bioinformatics↗
bioRxiv2017

Exploring thematic structure in 16S rRNA marker gene surveys

BackgroundAnalysis of microbiome data involves identifying co-occurring groups of taxa associated with sample features of interest (e.g., disease state). But elucidating key associations is often difficult since microbiome data are compositional, high dimensional, and sparse. Also, the configuration of co-occurring taxa may represent overlapping subcommunities that contribute to, for example, host status. Preserving the configuration of co-occurring microbes rather than detecting specific indicator species is more likely to facilitate biologically meaningful interpretations. In addition, analyses that utilize both taxonomic and predicted functional abundances typically independently characterize the taxonomic and functional profiles before linking them to sample information. This prevents investigators from identifying the specific functional components associate with which subsets of co-occurring taxa.\n\nResultsWe provide an approach to explore co-occurring taxa using \"topics\" generated via a topic model and then link these topics to specific sample classes (e.g., diseased versus healthy). Rather than inferring predicted functional content independently from taxonomic abundances, we instead focus on inference of functional content within topics, which we parse by estimating pathway-topic interactions through a multilevel, fully Bayesian regression model. We apply our methods to two large publically available 16S amplicon sequencing datasets: an inflammatory bowel disease (IBD) dataset from Gevers et al. and data from the American Gut (AG) project. When applied to the Gevers et al. IBD study, we demonstrate that a topic highly associated with Crohns disease (CD) diagnosis is (1) dominated by a cluster of bacteria known to be linked with CD and (2) uniquely enriched for a subset of lipopolysaccharide (LPS) synthesis genes. In the AG data, our approach found that individuals with plant-based diets were enriched with Lachnospiraceae, Roseburia, Blautia, and Ruminococcaceae, as well as fluorobenzoate degradation pathways, whereas pathways involved in LPS biosynthesis were depleted.\n\nConclusionsWe introduce an approach for uncovering latent thematic structure in the context of sample features for 16S rRNA surveys. Using our topic-model approach, investigators can (1) capture groups of co-occurring taxa termed topics, (2) uncover within-topic functional potential, and (3) identify gene sets that may guide future inquiry. These methods have been implemented in a freely available R package https://github.com/EESI/themetagenomics.

bioinformatics↗
bioRxiv2017

Opportunities And Obstacles For Deep Learning In Biology And Medicine

Deep learning, which describes a class of machine learning algorithms, has recently showed impressive results across a variety of domains. Biology and medicine are data rich, but the data are complex and often ill-understood. Problems of this nature may be particularly well-suited to deep learning techniques. We examine applications of deep learning to a variety of biomedical problems--patient classification, fundamental biological processes, and treatment of patients--and discuss whether deep learning will transform these tasks or if the biomedical sphere poses unique challenges. We find that deep learning has yet to revolutionize or definitively resolve any of these problems, but promising advances have been made on the prior state of the art. Even when improvement over a previous baseline has been modest, we have seen signs that deep learning methods may speed or aid human investigation. More work is needed to address concerns related to interpretability and how to best model each problem. Furthermore, the limited amount of labeled data for training presents problems in some domains, as do legal and privacy constraints on work with sensitive health records. Nonetheless, we foresee deep learning powering changes at both bench and bedside with the potential to transform several areas of biology and medicine.

bioinformatics↗