arXiv:2512.03158v1 Announce Type: new 
Abstract: Wastewater-based genomic surveillance has emerged as a powerful tool for population-level viral monitoring, offering comprehensive insights into circulating viral variants across entire communities. However, this approach faces significant computational challenges stemming from high sequencing noise, low viral coverage, fragmented reads, and the complete absence of labeled variant annotations. Traditional reference-based variant calling pipelines struggle with novel mutations and require extensive computational resources. We present a comprehensive framework for unsupervised viral variant detection using Vector-Quantized Variational Autoencoders (VQ-VAE) that learns discrete codebooks of genomic patterns from k-mer tokenized sequences without requiring reference genomes or variant labels. Our approach extends the base VQ-VAE architecture with masked reconstruction pretraining for robustness to missing data and contrastive learning for highly discriminative embeddings. Evaluated on SARS-CoV-2 wastewater sequencing data comprising approximately 100,000 reads, our VQ-VAE achieves 99.52% mean token-level accuracy and 56.33% exact sequence match rate while maintaining 19.73% codebook utilization (101 of 512 codes active), demonstrating efficient discrete representation learning. Contrastive fine-tuning with different projection dimensions yields substantial clustering improvements: 64-dimensional embeddings achieve +35% Silhouette score improvement (0.31 to 0.42), while 128-dimensional embeddings achieve +42% improvement (0.31 to 0.44), clearly demonstrating the impact of embedding dimensionality on variant discrimination capability. Our reference-free framework provides a scalable, interpretable approach to genomic surveillance with direct applications to public health monitoring.

تم تطوير إطار جديد يستخدم مشفرات متغيرة الكمية (VQ-VAE) للكشف غير الخاضع للإشراف عن المتغيرات الفيروسية في تسلسل الجينوم لمياه الصرف الصحي، مما يعالج التحديات مثل الضوضاء العالية في التسلسل والتغطية الفيروسية المنخفضة. لا تعتمد هذه الطريقة على الجينومات المرجعية أو تسميات المتغيرات، مما يعزز القدرة على مراقبة الفيروسات على مستوى السكان.

Se ha desarrollado un nuevo marco que utiliza autoencoders variacionales cuantificados por vectores (VQ-VAE) para la detección no supervisada de variantes virales en el secuenciamiento genómico de aguas residuales, abordando desafíos como el alto ruido de secuenciación y la baja cobertura viral. Este método no depende de genomas de referencia ni de variantes etiquetadas, mejorando la capacidad de monitoreo viral a nivel poblacional.

Un nouveau cadre utilisant des autoencodeurs variationnels quantifiés par vecteurs (VQ-VAE) a été développé pour la détection non supervisée des variants viraux dans le séquençage génomique des eaux usées, répondant à des défis tels que le bruit de séquençage élevé et la faible couverture virale. Cette méthode ne repose pas sur des génomes de référence ou des variants étiquetés, améliorant ainsi la capacité de surveillance virale au niveau de la population.

A new framework utilizing Vector-Quantized Variational Autoencoders (VQ-VAE) has been developed for unsupervised viral variant detection in wastewater genomic sequencing, addressing challenges such as high sequencing noise and low viral coverage. This method does not rely on reference genomes or labeled variants, enhancing the capability for population-level viral monitoring.

Contrastive Deep Learning for Variant Detection in Wastewater Genomic Sequencing

arXiv:2512.06609v1 Announce Type: new 
Abstract: Vector quantized variational autoencoder (VQ-VAE) is a discrete auto-encoder that compresses images into discrete tokens. It is difficult to train due to discretization. In this paper, we propose a simple yet effective technique, dubbed Gaussian Quant (GQ), that converts a Gaussian VAE with certain constraint into a VQ-VAE without training. GQ generates random Gaussian noise as a codebook and finds the closest noise to the posterior mean. Theoretically, we prove that when the logarithm of the codebook size exceeds the bits-back coding rate of the Gaussian VAE, a small quantization error is guaranteed. Practically, we propose a heuristic to train Gaussian VAE for effective GQ, named target divergence constraint (TDC). Empirically, we show that GQ outperforms previous VQ-VAEs, such as VQGAN, FSQ, LFQ, and BSQ, on both UNet and ViT architectures. Furthermore, TDC also improves upon previous Gaussian VAE discretization methods, such as TokenBridge. The source code is provided in https://github.com/tongdaxu/VQ-VAE-from-Gaussian-VAE.

تم تقديم تقنية جديدة تُدعى Gaussian Quant (GQ) لتحسين تدريب مشفرات التباين الكمي المتجهة (VQ-VAE) التي تُستخدم لضغط الصور إلى رموز منفصلة. تتيح هذه الطريقة تحويل VAE غاوسي إلى VQ-VAE دون الحاجة إلى تدريب مكثف، مما يبسط العملية ويحسن الأداء.

Se ha introducido una nueva técnica llamada Gaussian Quant (GQ) para mejorar el entrenamiento de los Autoencoders Variacionales Cuantificados por Vectores (VQ-VAE), que se utilizan para comprimir imágenes en tokens discretos. Este método permite convertir un VAE gaussiano en un VQ-VAE sin necesidad de un entrenamiento extenso, simplificando así el proceso y mejorando el rendimiento.

Une nouvelle technique appelée Gaussian Quant (GQ) a été introduite pour améliorer l'entraînement des Autoencodeurs Variationnels à Quantification Vectorielle (VQ-VAE), utilisés pour compresser des images en tokens discrets. Cette méthode permet de convertir un VAE gaussien en VQ-VAE sans nécessiter d'entraînement intensif, simplifiant ainsi le processus et améliorant les performances.

A new technique called Gaussian Quant (GQ) has been introduced to enhance the training of Vector Quantized Variational Autoencoders (VQ-VAE), which are used for compressing images into discrete tokens. This method allows for the conversion of a Gaussian VAE into a VQ-VAE without the need for extensive training, thereby simplifying the process and improving performance.

Contrastive Deep Learning for Variant Detection in Wastewater Genomic Sequencing

Was this article worth reading? Share it

LucidQuery AI

Videotok

SVGenius