The Finnic Peoples of Russia: Genetic Structure Inferred from Genome-Wide and Y-Chromosome Data.

Eastern Finnic populations, including Karelians, Veps, Votes, Ingrians, and Ingrian Finns, are a significant component of the history of Finnic populations, which have developed over ~3 kya. Yet, these groups remain understudied from a genetic point of view. In this work, we explore the gene pools of Karelians (Northern, Tver, Ludic, and Livvi), Veps, Ingrians, Votes, and Ingrian Finns using Y-chromosome markers (N = 357) and genome-wide autosomes (N = 67) and in comparison with selected Russians populations of the area (N = 763).

The significance of monitoring respiratory sample cultures and polymerase chain reaction tests for detecting bacterial pathogens in severely and critically ill patients with COVID-19.

Background: Bacterial superinfection is one of the most common and potentially lethal complications in severely and critically ill patients with COVID-19.

Objectives: To determine the colonisation time frame and the spectrum of potential bacterial pathogens in respiratory samples from patients with severe and critical COVID-19, using routine culture and polymerase chain reaction (PCR) tests.

Methods: A prospective observational study was conducted on patients aged ≥18 years with confirmed severe and critical COVID-19 who were admitted to or transferred to the intensive care unit (ICU).

QuIP: 2-Bit Quantization of Large Language Models With Guarantees.

This work studies post-training parameter quantization in large language models (LLMs). We introduce quantization with incoherence processing (QuIP), a new method based on the insight that quantization benefits from weight and Hessian matrices, i.e.

Online Calibrated and Conformal Prediction Improves Bayesian Optimization.

Accurate uncertainty estimates are important in sequential model-based decision-making tasks such as Bayesian optimization. However, these estimates can be imperfect if the data violates assumptions made by the model (e.g.

Cross-species modeling of plant genomes at single nucleotide resolution using a pre-trained DNA language model.

Interpreting function and fitness effects in diverse plant genomes requires transferable models. Language models (LMs) pre-trained on large-scale biological sequences can learn evolutionary conservation and offer cross-species prediction better than supervised models through fine-tuning limited labeled data. We introduce PlantCaduceus, a plant DNA LM based on the Caduceus and Mamba architectures, pre-trained on a curated dataset of 16 Angiosperm genomes.