Datasets for Bumps Paper
========================

Ubiquitin
---------

* NMR/ folder contains all experimental data in TopSpin formats, including VA lists and pulse sequences.
* ubiquitin_integrals.pickle contains all extracted integrals from the experimental datasets. This can be read in Python3 as

```
import pickle
import numpy as np

with open('ubiquitin_integrals.pickle', 'rb') as f:
    data = pickle.load(f)
```

This contains a dictionary, where each key represents a specific peak. data[peak] is then a dictionary containing

  - 'H', 'N' - chemical shifts, in pts (relative to spectrum) and in ppm.
  - 'rates' - all full relaxation rate measurements for that peak. For each rate measurement, there are:
    - 'time_s' - list of time points
    - 'amplitude' - peak intensity
    - 'noise' - estimate of uncertainty in peak intensity at 1 std
    - 'theta' - calculated theta value in radians
    - 'w1_kHz' - effective nutation frequency, corrected for offset.
    - 'R1p', 'R1pe' - fit R1p and R1pe values (note that the following analysis script recalculates these
    - 'overlap' - true/false depending whether the peak shows overlap with other sites
  - 'profiles' - all full fast mode relaxation profiles.
    - 'w1_kHz' - effective nutation frequency at this site
    - 'theta_rad' - theta value for each frequency
    - 'amplitude'/'noise' - peak intensity and estimated uncertainty
    - 'overlap' - true/false depending whetehr the peak shows overlap with other sites.
  - 'assignment' - assignment. In the case that the peak shows overlap, there may be multiple assignments here.

* analysis_ubiquitin.py reads in the  experimental integrals in ubiquitin_integrals.pickle and performs the analysis to extract R1p from these. This then outputs 'ubiquitin_rates.pickle'
* ubiquitin_rates.pickle contains a list. Each element in the list is a tuple of:

  - the first assignment in the assignment list
  - the corresponding entry in the integrals data structure
  - tuples (w1, R1p, R1pe) for each:
    - 100 kHz HCN 700 MHz
    - 100 kHz DCN 700 MHz
    - 55 kHz DCN 700 MHz
    - 55 kHz DCN 700 MHz @ 108 ppm offset *
    - 55 kHz DCN 700 MHz @ 113 ppm offset *
    - 55 kHz DCN 700 MHz @ 118 ppm offset *
    - 55 kHz DCN 700 MHz @ 123 ppm offset *
    - 55 kHz DCN 700 MHz @ 128 ppm offset *
    - 55 kHz DCN 600 MHz 
    - 100 kHz HCN 700 MHz with CW decoupling
    - 100 kHz DCN 700 MHz with CW decoupling
    (the entries marked with * were recorded and analysed, but not included in the paper. They are included here for the interested reader)
    
* Schanda2010_850_R1.txt, R1 relaxation rates at 850 MHz as published in Schanda et al 2010.
* local_1H1D/ 
  * BMRB 68 - dataset from BMRB entry 68, containing 1H chemical shifts for ubiquitin in solution
  * 3ons_contacts.pdb - structure of ubiquitin with nearest contacts, with protons added using ChimeraX.
  * generate.py - script which implements the 4 parameter model of the bumps based on structure and shifts.

TET2
----

* NMR/ folder contains all experimental data in TopSpin formats, including VA lists and pulse sequences.
* TET_integrals.pickle contains all extracted integrals from the experimental datasets.
* analysis_TET.py analyses the integrals to output
* TET_rates.pickle: as in ubiquitin, with 
  - the first assignment in the assignment list
  - the corresponding entry in the integrals data structure
  - tuples (w1, R1p, R1pe) for each:
    - 55 kHz DCN 700 MHz 
    - 55 kHz DCN 700 MHz with CW decoupling