arXiv subjects

Claudia Solís-Lemus

Publications and source records attributed to Claudia Solís-Lemus.

7 recordsLinked to original sources

arXiv2026

MAST: Label-Efficient, Robust, and Generalizable Sound Detection for Biodiversity Monitoring via Masked Audio Pretraining and Self-Training

Passive acoustic monitoring can measure biodiversity at larger scales, but time--frequency annotation of animal vocalizations is expensive, site-specific, and difficult to sustain at scale. We present a label-efficient sound detection framework that combines masked audio pretraining with a lightweight detector on mel spectrograms, then further improves robustness through iterative self-training on unlabeled audio. We first pretrain a ViT-based encoder on unlabeled recordings via masked reconstruction and transfer the encoder to a detection backbone. To better separate animal sounds from confounding background, we add a box-level contrastive loss that pulls matched event regions together while pushing noisy negatives apart. We then apply a two-stage pseudo-labeling curriculum to exploit large unlabeled pools without additional annotation. We evaluate the performance on two ecologically distinct domains: tropical rainforest soundscapes (Indonesia) and bird vocalizations in Mediterranean habitats (Spain). On both domains, masked audio pretraining and contrastive learning consistently improve time--frequency detection under temporal and cross-site distribution shift, and self-training yields further gains in out-of-distribution performance. On the rainforest domain, MAST with self-training achieves +0.22 mAP and +0.24 F1 over the strongest baseline under cross-site shift. On the bird domain, self-training achieves +0.12 mAP and +0.10 F1 over the strongest baseline under cross-site shift. Overall, our results show that MAST can effectively extend self-supervised audio representations from clip-level tasks to robust box-level localization across diverse bioacoustic settings, providing a practical path for biodiversity monitoring with limited labels.

cs.SD↗

arXiv2026

Generating the Unheard: Phylogeny-Guided Latent Generation for Ancestral Sound Reconstruction

What did an ancestral bird species sound like? Existing ancestral state reconstruction methods can infer low-dimensional traits such as morphological characters at internal nodes of a phylogenetic tree, but no one has tried to produce rich perceptual signals such as audio. Some of the challenges include inferred representations that are either too low-dimensional to decode or lie in non-generative feature spaces, so no method to date can produce ancestral audio. We introduce the first framework that generates plausible ancestral vocalizations. Our pipeline encodes bird recordings into a VAE latent space, learns a low-dimensional trait projection aligned with phylogenetic distances, performs ancestral inference in this trait space, and recovers decodable latents through an anchored inverse lift before emitting novel waveforms for each ancestral node. Because the entire pipeline stays within a decodable latent space, every internal node receives a genuinely new audio output representing plausible intermediate ancestral sounds unavailable to retrieval-based alternatives. Experiments on two phylogenetically distant bird clades, 21-species Tyrannidae and 19-species Paridae, show that our method is the only approach that simultaneously achieves genuine generation, phylogenetic consistency, and naturalistic audio quality across both datasets.

cs.SD↗

arXiv2026

Matrix representations and distance metrics for unlabeled ranked phylogenetic networks

Phylogenetic networks are graphs inferred from molecular sequence data that represent ancestral histories shaped by reticulate processes such as recombination, hybridization, and horizontal gene transfer. We introduce a family of distance metrics for rooted, ranked, unlabeled phylogenetic networks, extending a previously developed distance for ranked trees. Our approach relies on a bijective triangular matrix representation of phylogenetic networks that captures the temporal order of internal events, speciations, and hybridizations. Our metrics, defined as standard matrix norms, allow efficient quantitative comparisons of network topologies, timed networks and networks with differing numbers of hybridizations. Our distance can be used for both isochronous networks where all tips are sampled at one time point, and heterochronous networks where tips are allowed to be sampled at different time points. We show that our metrics capture biologically meaningful differences among evolutionary histories in both simulations and empirical posterior distributions of viral phylogenetic networks. These tools fill a methodological gap, enabling principled comparisons of ranked, unlabeled phylogenetic networks, including ancestral recombination graphs.

stat.ME↗

arXiv2025

MiNAA-WebApp: A Web-Based Tool for the Visualization and Analysis of Microbiome Networks

Microbial networks, representing microbes as nodes and their interactions as edges, are crucial for understanding community dynamics in various environments. Analyzing microbiome networks is crucial for identifying keystone taxa that play central roles in maintaining microbial community structure and function, assessing how environmental changes such as pollution, climate shifts, or land use affect microbial dynamics, tracking disease progression by revealing alterations in microbial interactions over time, and predicting microbial community responses to interventions such as antibiotics, probiotics, or changes in diet and habitat. The complexity of microbial interactions necessitates the use of computational tools such as the MiNAA-WebApp, available at https://minaa.wid.wisc.edu, which enhances the accessibility of the Microbiome Network Alignment Algorithm MiNAA. This tool allows researchers to align microbial networks and explore ecological relationships and community dynamics without extensive computational skills. Originally, MiNAA's command-line interface limited its usability for those without programming backgrounds. The web-based MiNAA-WebApp addresses this shortcoming by offering an intuitive interface with visualization tools, allowing easy exploration and analysis of microbial networks. The web app is designed for microbiome networks but also applicable to other biological networks, broadening its use in computational biology and making network-based research accessible to a wider audience.

q-bio.MN↗

arXiv2024

Estimating sparse direct effects in multivariate regression with the spike-and-slab LASSO

The multivariate regression interpretation of the Gaussian chain graph model simultaneously parametrizes (i) the direct effects of $p$ predictors on $q$ outcomes and (ii) the residual partial covariances between pairs of outcomes. We introduce a new method for fitting sparse Gaussian chain graph models with spike-and-slab LASSO (SSL) priors. We develop an Expectation Conditional Maximization algorithm to obtain sparse estimates of the $p \times q$ matrix of direct effects and the $q \times q$ residual precision matrix. Our algorithm iteratively solves a sequence of penalized maximum likelihood problems with self-adaptive penalties that gradually filter out negligible regression coefficients and partial covariances. Because it adaptively penalizes individual model parameters, our method is seen to outperform fixed-penalty competitors on simulated data. We establish the posterior contraction rate for our model, buttressing our method's excellent empirical performance with strong theoretical guarantees. Using our method, we estimated the direct effects of diet and residence type on the composition of the gut microbiome of elderly adults.

stat.ME↗

arXiv2024

Human Limits in Machine Learning: Prediction of Plant Phenotypes Using Soil Microbiome Data

The preservation of soil health is a critical challenge in the 21st century due to its significant impact on agriculture, human health, and biodiversity. We provide the first deep investigation of the predictive potential of machine learning models to understand the connections between soil and biological phenotypes. We investigate an integrative framework performing accurate machine learning-based prediction of plant phenotypes from biological, chemical, and physical properties of the soil via two models: random forest and Bayesian neural network. We show that prediction is improved when incorporating environmental features like soil physicochemical properties and microbial population density into the models, in addition to the microbiome information. Exploring various data preprocessing strategies confirms the significant impact of human decisions on predictive performance. We show that the naive total sum scaling normalization that is commonly used in microbiome research is not the optimal strategy to maximize predictive power. Also, we find that accurately defined labels are more important than normalization, taxonomic level or model characteristics. In cases where humans are unable to classify samples accurately, machine learning model performance is limited. Lastly, we provide domain scientists via a full model selection decision tree to identify the human choices that optimize model prediction power. Our work is accompanied by open source reproducible scripts (https://github.com/solislemuslab/soil-microbiome-nn) for maximum outreach among the microbiome research community.

stat.ML↗

arXiv2016

Inferring phylogenetic networks with maximum pseudolikelihood under incomplete lineage sorting

Phylogenetic networks are necessary to represent the tree of life expanded by edges to represent events such as horizontal gene transfers, hybridizations or gene flow. Not all species follow the paradigm of vertical inheritance of their genetic material. While a great deal of research has flourished into the inference of phylogenetic trees, statistical methods to infer phylogenetic networks are still limited and under development. The main disadvantage of existing methods is a lack of scalability. Here, we present a statistical method to infer phylogenetic networks from multi-locus genetic data in a pseudolikelihood framework. Our model accounts for incomplete lineage sorting through the coalescent model, and for horizontal inheritance of genes through reticulation nodes in the network. Computation of the pseudolikelihood is fast and simple, and it avoids the burdensome calculation of the full likelihood which can be intractable with many species. Moreover, estimation at the quartet-level has the added computational benefit that it is easily parallelizable. Simulation studies comparing our method to a full likelihood approach show that our pseudolikelihood approach is much faster without compromising accuracy. We applied our method to reconstruct the evolutionary relationships among swordtails and platyfishes ($Xiphophorus$: Poeciliidae), which is characterized by widespread hybridizations.

q-bio.PE↗