Keith A. Crandall

Mobile DNA Activity in Parkinson's Disease: A Locus-Specific View of Endogenous Retroviruses

This study identifies locus-specific HERV expression patterns in Parkinson's disease patients, revealing shared retroviral signals and immune-related pathway dysregulation across different PD subtypes.

seqLens: optimizing language models for genomic predictions

This study develops seqLens, a genomic language model family, and evaluates its performance on phenotypic predictions using diverse pretraining datasets and tokenization strategies.

resLens: genomic language models to enhance antibiotic resistance gene detection

resLens improves antibiotic resistance gene detection by using genomic language models trained on curated datasets, outperforming reference-based methods in varied scenarios.

Phylogenetically-informed crayfish conservation in the face of climate change

This study identifies priority crayfish species for conservation based on phylogenetic diversity and climate change vulnerability, finding that protecting these species could safeguard significant evolutionary history.

Guided tokenization and domain knowledge enhance genomic language models' performance

This paper introduces guided tokenization and domain adaptation to improve genomic language models' performance in classification and annotation tasks.

Curating 16S rRNA databases enhances taxonomic accuracy and computational efficiency in microbial profiling

The paper curates 16S rRNA databases to improve taxonomic accuracy and computational efficiency in microbial profiling by cleaning sequences and reducing redundancy.

Discovering genotype-phenotype relationships with machine learning and the Visual Physiology Opsin Database (VPOD)

The paper introduces a database of opsin genotypes and λmax phenotypes and demonstrates regression-based ML models can predict λmax and identify functionally critical sites.

Metagenomic profiling pipelines improve taxonomic classification for 16S amplicon sequencing data

The paper evaluates taxonomic classification accuracy of 16S amplicon sequencing data using different pipelines and reference databases, finding that whole-genome tools outperform 16S-specific ones.

Diversity and genomic determinants of the microbiomes associated with COVID-19 and non-COVID respiratory diseases

This study compares microbiome diversity and genomic features in COVID-19, COPD, and URTI samples to identify differences in composition and functional pathways.

COVID-19 Biomarkers in research: Extension of the OncoMX cancer biomarker data model to capture biomarker data from other diseases.

The paper extends a cancer biomarker data model to create a COVID-19 biomarker knowledgebase, integrating 138 biomarkers and 265 type combinations with ontological annotations.

Evaluation of haplotype callers for next-generation sequencing of viruses

This study evaluates twelve haplotype callers for NGS viral sequencing, finding they perform poorly under high genetic diversity, with PredictHaplo and PEHaplo showing better accuracy for HIV-like diversity levels.

Schizophrenia risk from locus-specific human endogenous retroviruses

The paper identifies 148 HERVs associated with schizophrenia risk using comprehensive annotation and RNA-sequencing data from brain samples.

Telescope: Characterization of the retrotranscriptome by accurate estimation of transposable element expression

Telescope accurately estimates transposable element expression in RNA-seq data by resolving ambiguously mapped fragments using a Bayesian model, revealing cell-type specific retrotranscriptome differences.

Association of milk microbiome in bovine clinical mastitis and their functional implications in cows in Bangladesh

The paper analyzes milk microbiomes from bovine clinical mastitis and healthy samples in Bangladesh, identifying microbial and functional differences associated with the disease.

ReQTL - an allele-level measure of variation-expression genomic relationships

The paper introduces ReQTL, an eQTL analysis method using RNA genetic variation instead of DNA genotypes, showing high performance in identifying molecular associations from RNA-sequencing data.