Skip to content

Relaxation & dynamics

Two layers here: pull deposited relaxation straight out of an entry with NMRStarEntry, or assemble it into a per-residue dynamics profile with RelaxationProfile, in the style of RelaxDB (Wayment-Steele, El Nesr et al.).

Deposited data, straight from an entry

import makeshift as ms

entry = ms.NMRStarEntry.from_bmrb(25013)
entry.datasets()                # which data types the entry holds
entry.relaxation("T2")          # R2 — also "T1"/"R1", "T1rho", "NOE"; units-aware
entry.order_parameters()        # model-free S2 (S2, Tau_e, Rex)

# Anything without a dedicated method:
entry.data_loop("spectral_density_values", "_Spectral_density")

Building a dynamics profile

RelaxationProfile aligns R1/R2/NOE to the sequence, forms the R₂/R₁ observable, compares it to a HYDRONMR rigid-body prediction, and labels each residue by motional regime.

from makeshift.relaxation import RelaxationProfile

prof = RelaxationProfile.from_bmrb(25013)   # pulls T1/T2/NOE, aligns to sequence
prof.add_rigid_prediction()                 # runs HYDRONMR on a structure
print(prof.label())                         # per-residue motion string
prof.plot("R2_R1")

You can also build from an already-parsed entry with RelaxationProfile.from_entry(entry).

The table

prof.table has one row per sequence position (1-indexed), with columns Seq_ID, residue, R1, R1_err, R2, R2_err, NOE, NOE_err, R2_R1, R2_R1_err, and has_data — plus scaled_R2_R1_pred and label once the rigid-prediction and labelling steps have run.

The rigid-body prediction

add_rigid_prediction() runs HYDRONMR on a structure and scales its rigid R₂/R₁ to the data. The structure can be:

  • a local PDB file — add_rigid_prediction("model.pdb")
  • a PDB id (fetched from RCSB) — add_rigid_prediction("1WRP")
  • a UniProt accession (fetched from AlphaFold DB) — add_rigid_prediction("P0DP23")

With no argument it uses the first available PDB or AlphaFold model based on the entry cited.

makeshift does not predict structure

It only fetches an existing experimental or predicted model. See Datasets & structures for more information or provide your own prediction.

Motion labels

Labels for each residue by motion regime against a HYDRONMR rigid-body prediction — this follows the RelaxDB curation described in (Wayment-Steele, El Nesr et al)(https://www.biorxiv.org/content/10.1101/2025.03.19.642801v3).

label() assigns one token per residue and returns the label string:

Token Meaning
A ordered
^ µs–ms exchange (elevated R₂/R₁)
v ps–ns motion (hetNOE ≤ 0.65)
b both
. peak missing
t disordered terminus
p proline

Plotting

prof.plot("R2_R1")   # any observable column, along the sequence

Full API

See the Relaxation reference. The CPMG relaxation-dispersion pipeline has its own guide.