NERD
NERD (Nucleic acid Energetics from Reactivity Data) is a unified toolkit for extracting nucleotide-level thermodynamic and kinetic information from chemical probing experiments. It streamlines sample registration, mutation-count processing, and kinetic/melt-fit analysis, providing an organized path from raw sequencing data to time-course and temperature-gradient energetics.
Quick Start
# create a virtual environment (recommended)
python -m venv .venv
source .venv/bin/activate
# install NERD
pip install nerd-pipeline
# initialize a project from an example config
nerd run create demo_folder/01_create_samples/configs/create_meta.yaml
# execute a probe timecourse fit
nerd run probe_timecourse demo_folder/05_probe_tc_kinetics/configs/probe_tc.yaml
Key Features
- Unified SQLite backbone linking constructs, buffers, reactions, fits, and artifacts
- CLI-based execution with consistent logging, caching, and reproducibility guarantees
- Pluggable fitting engines for probe timecourses, Arrhenius, and two-state melt models
- Rich filtering (construct, buffer, probe, nt_id, base) and outlier management in configs
- Automated metadata ingestion from CSVs and lab notebooks
- Portable outputs (JSON artifacts, logs, database entries) for downstream visualization
- Local and remote execution modes (local CPU, SLURM, SSH)
- Tutorial notebooks and guides for sample organization, NMR kinetics, probe analyses, and tempgrad fits
Documentation Map
- Getting Started: Sample organization & database layout
- Guides:
- Probe timecourse workflow
- NMR kinetics workflow
- Temperature-gradient fitting
- CLI Reference:
nerd run … - create
- mut_count
- nmr_create
- nmr_deg_kinetics
- nmr_add_kinetics
- probe_timecourse
- tempgrad_fit
- Configuration & Workflow Guide: See guides above plus example YAML in
demo_folder/ - Architecture Overview: Sample organization & database layout
- Examples: Explore
demo_folder/configs,examples/nerd.sqlite, and accompanying notebooks
How It Works
[Pipeline diagram here: sample CSVs & raw data → NERD CLI tasks → SQLite database → fit engines → JSON/plots]
- Register constructs, buffers, reaction groups, and traces with
nerd run create. - Quantify sequencing or NMR data via task-specific runners (
mut_count,nmr_*). - Execute probe timecourse fits to derive kinetic parameters per nucleotide.
- Run
tempgrad_fitto combine kinetics across temperatures with Arrhenius or two-state models. - Visualize results using bundled notebooks or your own data science stack.
Example Output (Optional)
manuscript_pipeline/05_probe_tempgrad_fit/tempgrad_fit/latest/results/tempgrad_result.json
Citation
Please cite the NERD pipeline as: [placeholder—add citation here].