1kx8 readme

Author: jaumebonet

03.experimental/1kx8/README.ipynb

@@ -0,0 +1,221 @@
+
+
+```python
+# Default Libraries
+import os
+import StringIO
+
+# External Libraries
+import pandas as pd
+import numpy as np
+import matplotlib
+import matplotlib.pyplot as plt
+import seaborn as sns
+import networkx as nx
+from IPython.display import FileLink, FileLinks
+
+# Own Libraries
+import rstoolbox  # Learn about this library at: http://lpdi-epfl.github.io/RosettaSilentToolbox
+
+import readme
+
+# Global Parameters
+plt.rcParams['svg.fonttype'] = 'none'
+rstoolbox.utils.format_Ipython()
+```
+
+
+
+
+<style>
+table.dataframe {
+    font-family: monospace;
+}
+</style>
+
+
+
+# Design of a RSVF site II carrier from 1KX8
+
+> This summary does not track all attempts and variations, just the pathway towards obtaining the designs that were experimentally characterized.
+
+
+```python
+# Direct information obtain from scoring 1kx8 with Rosetta
+baseline = rstoolbox.io.get_sequence_and_structure(os.path.join('selection', 'baseline', '1kx8.pdb'))
+baseline
+```
+
+
+
+
+<div>
+<style scoped>
+    .dataframe tbody tr th:only-of-type {
+        vertical-align: middle;
+    }
+
+    .dataframe tbody tr th {
+        vertical-align: top;
+    }
+
+    .dataframe thead th {
+        text-align: right;
+    }
+</style>
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>score</th>
+      <th>fa_atr</th>
+      <th>fa_rep</th>
+      <th>fa_sol</th>
+      <th>fa_intra_rep</th>
+      <th>fa_intra_sol_xover4</th>
+      <th>lk_ball_wtd</th>
+      <th>fa_elec</th>
+      <th>pro_close</th>
+      <th>hbond_sr_bb</th>
+      <th>hbond_lr_bb</th>
+      <th>hbond_bb_sc</th>
+      <th>hbond_sc</th>
+      <th>dslf_fa13</th>
+      <th>omega</th>
+      <th>fa_dun</th>
+      <th>p_aa_pp</th>
+      <th>yhh_planarity</th>
+      <th>ref</th>
+      <th>rama_prepro</th>
+      <th>time</th>
+      <th>description</th>
+      <th>sequence_A</th>
+      <th>structure_A</th>
+      <th>phi_A</th>
+      <th>psi_A</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>165.939</td>
+      <td>-594.399</td>
+      <td>185.277</td>
+      <td>426.021</td>
+      <td>1.376</td>
+      <td>25.587</td>
+      <td>-16.933</td>
+      <td>-141.967</td>
+      <td>7.165</td>
+      <td>-50.138</td>
+      <td>-2.655</td>
+      <td>-6.671</td>
+      <td>-6.171</td>
+      <td>-2.659</td>
+      <td>0.059</td>
+      <td>326.333</td>
+      <td>-8.581</td>
+      <td>0.0</td>
+      <td>16.661</td>
+      <td>7.633</td>
+      <td>2.0</td>
+      <td>1kx8_0001</td>
+      <td>NINLDEILANKRLLVAYVNCVMERGKCSPEGKELKEHLQDAIENGCKKCTENQEKGAYRVIEHLIKNEIEIWRELTAKYDPTGNWRKKYEDRAKAAGIVI</td>
+      <td>LLLLHHHHHLHHHHHHHHHHHLLLLLLLHHHHHHHHHHHHHHHHLLLLLLHHHHHHHHHHHHHHHHHLHHHHHHHHHHHLLLLLLHHHHHHHHHHHLLLL</td>
+      <td>[0.0, -46.7858, -126.674, ...]</td>
+      <td>[-42.2254, 137.85, -165.222, ...]</td>
+    </tr>
+  </tbody>
+</table>
+</div>
+
+
+
+## Making the designs
+
+The construction of the **1kx8**-derived designs implies the grafting to RSVF's site II as represented in [3ixt.pdb](design/3ixt.pdb) without its neutralizing antibody.
+
+The process was performed in a two-generation step.
+
+### Step 1
+
+Site II was grafted into **1kx8** by means of the _NubInitioMover_ and the sequence space after folding was explored with _FastDesign_. All within the RosettaScript [fold_design.xml](design/fold_design.xml) into a [SLURM](https://slurm.schedmd.com/)-based cluster through the submiter script [submit_gen1.sh](design/submit_gen1.sh).
+
+A total of 12483 decoys were generated. The aim of this firts run was to explore and find a folded conformation that would:
+
+* carry and stabilize at backbone level site II.
+* fold back to a conformation close to **1kx8** keeping the proper beta pairing.
+
+From all the decoys, [188_1kx8_0033](design/188_1kx8_0033.pdb) was selected as seed for the next step of the process. The values of this decoy in the context of the design population are marked with a red line.
+
+
+```python
+step1 = rstoolbox.io.parse_rosetta_file('design/1kx8_silent.minisilent.gz')
+readme.plot_step1(step1)
+```
+
+
+![png](README_files/README_4_0.png)
+
+
+### Step 2
+
+The second generation derives from [188_1kx8_0033](design/188_1kx8_0033.pdb) and basically runs the same conditions but using this decoy instead of **1kx8** as template.
+
+The executed script is still  [fold_design.xml](design/fold_design.xml), but the submiter script is [submit_gen2.sh](design/submit_gen2.sh).
+
+A total of 12493 decoys were generated.
+
+From all the decoys, the top 500 best scorers were evaluated in terms of packing, BUNS and secondary structure divergence form the starting decoy with the [fastrelax_cav_vol_buns.xml](design/fastrelax_cav_vol_buns.xml) RosettaScript after global relaxation. The six decoys with the best behaviour after evaluating this parameters and some visual inspection became the seed for the seven desings of **generation 1**. The values for those designs are highlighted in the distributions in red.
+
+
+```python
+step2 = rstoolbox.io.parse_rosetta_file('design/1kx8_silent2.minisilent.gz')
+readme.plot_step2(step2)
+```
+
+
+![png](README_files/README_6_0.png)
+
+
+
+```python
+step3 = rstoolbox.io.parse_rosetta_file('design/top500score.silent')
+readme.plot_step3(step3)
+```
+
+
+![png](README_files/README_7_0.png)
+
+
+### Generation 1
+
+The seven designs of **generation 1** derive from the 6 selected from **step 3** after human-guided design (mutations were selected and imposed manually).
+
+### Generation 2
+
+The eight designs of **generation 2** derive from disulfidized versions of **generation 1** with automatic mutations added to fit the disulfides.
+
+
+Data from those two generations is discussed in the paper.
+
+# Experimental Data
+
+
+```python
+designs = readme.load_experimental()
+readme.firstgen_plot(designs)
+```
+
+
+![png](README_files/README_9_0.png)
+
+
+
+```python
+readme.secondgen_plot(designs)
+```
+
+
+![png](README_files/README_10_0.png)
+