Software

Variant interpretation

RegulomeDB, SURF, TURF, and TLand

RegulomeDB annotates variants with functional-genomics evidence and predictive models to help interpret noncoding regulatory variation. RegulomeDB v2 added expanded functional-genomics data, improved scoring, tissue-aware prediction resources, and visualization support. TURF and TLand extend this framework to prioritize regulatory variants in tissue-, organ-, and cell-specific contexts.

References: Boyle et al., Genome Research, 2012; Dong et al., Nature Genetics, 2023; Dong and Boyle, Nucleic Acids Research, 2021; Zhao, Dong, and Boyle, bioRxiv, 2023.

Quality control

ENCODE Blacklist

The ENCODE Blacklist identifies genomic regions that show anomalous signal across next-generation sequencing experiments. Removing these regions is an important quality-control step for ChIP-seq, DNase-seq, ATAC-seq, and related functional-genomics assays.

Reference: Amemiya, Kundaje, and Boyle, Scientific Reports, 2019.

Comparative genomics

Human–Mouse SOM Browser

Self-organizing map resources organize and compare genome-wide regulatory activity across human and mouse tissues and cell types. The SOM Browser provides an interactive resource, while the companion repository contains the comparative analysis code.

Reference: Diehl et al., Nucleic Acids Research, 2018.

Peak calling

F-Seq and F-Seq2

F-Seq is a feature-density estimator for identifying biologically meaningful signal-enriched regions from high-throughput sequencing data. F-Seq2 is a Python rewrite and extension that adds dynamic local statistics, support for common regulatory-genomics assays, and IDR-aware peak-calling workflows.

References: Boyle et al., Bioinformatics, 2008; Zhao and Boyle, NAR Genomics and Bioinformatics, 2021.

Chromatin accessibility

TRACE and TRACE_GPU

TRACE is a hidden Markov model for transcription-factor footprinting and motif matching using chromatin-accessibility data, including DNase-seq and ATAC-seq. TRACE_GPU accelerates core calculations, including emission-matrix generation and Viterbi decoding, on GPUs.

Reference: Ouyang and Boyle, Genome Research, 2020.

Sequence modeling

SEMpl, SEMplMe, and SEMPLR

SNP Effect Matrices model the effect of sequence variants on transcription-factor binding affinity. SEMpl is a command-line implementation, SEMplMe incorporates DNA methylation, and SEMPLR provides an R/Bioconductor interface for scoring genomic positions and variants.

References: Nishizaki et al., Bioinformatics, 2019; Nishizaki and Boyle, BMC Bioinformatics, 2022; Kenney et al., Bioinformatics, 2026.

Tandem repeats

HMMSTR

HMMSTR is a modified profile hidden Markov model for determining tandem-repeat copy number directly from raw long-read sequencing data. It is optimized for targeted sequencing experiments.

Reference: Van Deynze et al., Nucleic Acids Research, 2025.

Plasmid sequencing

OnRamp

OnRamp streamlines pooled plasmid validation using bulk plasmid sequencing. The Boyle Lab repository provides the web-enabled application, and the associated bulkPlasmidSeq repository contains the command-line workflow.

Reference: Mumm et al., Genome Research, 2023.

Targeted long-read sequencing

NanoPal and Cas9 enrichment pipelines

NanoPal and the associated Cas9 targeted-enrichment pipelines support long-read enrichment and analysis of mobile-element insertions. The workflows cover multiple L1HS, Alu, and SVA families and include guide-RNA design, cleavage-site analysis, and methylation-analysis scripts.

Reference: McDonald et al., Nature Communications, 2021.

Nanopore analysis

Minimera

Minimera detects foldback chimeras in Oxford Nanopore sequencing data using minimizers. It is distributed as command-line binaries and Singularity containers.