RotationNotebook.v1.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Rotation Notebook: Volume 1\n",
    "## Contents:\n",
    "* [April 4, 2016](#April-4,-2016) - Met with Tanja, suggestions on improvements and things to try with DDG RosettaScript Protocol<br>\n",
    "* [April 7, 2016](#April-7,-2016) - New directions while we are waiting for a repsonse from Colin\n",
    "* [April 10, 2016](#April-10,-2016) - Updates to DDG Backrub RosettaScript\n",
    "* [April 18, 2016](#April-18,-2016) - Suggestions for fixing RosettaScript Repack problem"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## April 4, 2016\n",
    "### Meeting with Tanja: \n",
    "<b>Suggestions:</b>\n",
    "* Look up minimization with constraints..\n",
    "* Parse my tag + minimize with constraint\n",
    "\n",
    "<b>Things to implement in the DDG RosettaScript Protocol:</b>\n",
    "* Repack with soft repulsive energy function (find soft version of talaris)\n",
    "* Add 1+/1- for all selected pivot resdiues in backrub mover\n",
    "* Figure out difference between backrub mover and protocol\n",
    "* Try less trials, more structures\n",
    "* Get backbone RMSD's and compare to WT structures\n",
    "* Make BackrunProtocol Mover for RosettaScripts\n",
    "* Minimize w/ constraints Mover vs. Application and get RMSD's\n",
    "* Add mute flag for MetropolisHastingsMover\n",
    "\n",
    "<b>Minimize_with_cst</b>\n",
    "* Harmonic All Ca - All Ca constraints\n",
    "* Increases fa_rep in steps [(0.1, 0.33, 1) * fa_rep] and minimizes at each step\n",
    "\n",
    "<b>RosettaScript Things to Test in the Future:</b>\n",
    "* AddConstraintsToCurrentConformationMover\n",
    "    * min_seq_sep - try different minimum sequence separation between pairs of constrained residues"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## April 7, 2016\n",
    "\n",
    "Talked with Kyle about the direction of my rotation so far and decided we should wait Colin's response before continuing with the DDG RosettaScript Protocol. The rationale is that my time will better spent working on something else instead of trying to implement the BackrubProtocol Mover in RosettaScripts on my own since Colin is going to help anyways.<br>\n",
    "\n",
    "<b>Other avenues to go down in the meantime:</b><br>\n",
    "\n",
    "* Testing different minimization methods on structures in the Zemu dataset\n",
    "    * The crappy (so I'm told) data from FirstZemuDataset_Workup where sereval DDG values blew up may be a result of the minimization method we used. The DDGmonomer app uses minimize_with_cst where I was using MinMover in RosettaScripts.\n",
    "    * Test minimize_with_cst app vs. MinMover in RosettaScripts on all structures in the Zemu dataset with two different minimization algorithms: lbfgs_armijo_nonmonotone (default) and dfpmin_armijo_nonmonotone.\n",
    "    * Compare output structure RMSDs\n",
    "<br><br>\n",
    "* Develop general method for correlate structural variation in protocol output (i.e. RMSD) with interesting factors (DDG scores, ensemble size, trial sizes, kT)\n",
    "    * How much backbone variation is required/useful?\n",
    "    * Input starting PDB, WT output PDBs, and mutant output PDBs\n",
    "    * All-vs-all RMSDs for any outputs (DDG WT, DDG mutants, minimization comparison structures)\n",
    "    * RMSD for 8A repack sphere (repacking) and overall structure\n",
    "    * Mutant Residue RMSD\n",
    "<br><br>\n",
    "* Find the PDB structures of Zemu dataset mutants\n",
    "    * [PyPDB] \n",
    "    * Some psuedo code for what I had in mind to try out:<br>\n",
    "    `    \n",
    "    for subdir in directory(data):\n",
    "        for files in subdir:`\n",
    "            mutlist = []\n",
    "            if file == 'mutations.resfile':\n",
    "                mut_type = mutant residue type\n",
    "                mut_pos = mutant residue position\n",
    "                mutlist.append([mut_type, mut_pos])\n",
    "            if file ends in '.pdb':\n",
    "                structure = parser.get_structure('TEST', data + subdir + files)\n",
    "                pdbcode = take the first four letters of filename.pdb\n",
    "            \n",
    "            #Find related structures using PyPDB module make_query()\n",
    "            MSQuery = pypdb.make_query(pdbcode, querytype = 'ModifiedStructuresQuery')\n",
    "            \n",
    "            #Obtain PDB sequence, introduce mutations, and search against PDB\n",
    "            ppb=CaPPBuilder()\n",
    "            for pp in ppb.build_peptides(structure): \n",
    "                print pp.get_sequence()\n",
    "            seq = polypeptide.get_sequence()\n",
    "            search_PDB_for_sequence(seq)\n",
    "\n",
    "1. Set up and submit runs for minimize_with_cst.default.linuxgccrelease and RosettaScripts with all Zemu Structures using lbfgs_armijo_nonmonotone (default) and dfpmin_armijo_nonmonotone\n",
    "2. Work on RMSD calculator for relevant variables\n",
    "3. Download structures for Zemu dataset mutants if time permits\n",
    "    \n",
    "[PyPDB]: https://github.com/williamgilpin/pypdb\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## April 10, 2016\n",
    "\n",
    "Backrub protocol basics:<br>\n",
    "* Start backrub MC\n",
    "    * If only sidechain rotamers is to be applied == True (P.rotatmer = 0.25):\n",
    "        * Pick a single residue and apply rotamers\n",
    "        * <b>Evaluate MC criterion and end Backrub MC step</b>\n",
    "    * If only sidechain rotamers is to be applied == False (P.rotatmer = 0.25):\n",
    "        * Apply backbone Backrub moves\n",
    "        * If only Backrub moves==True (P.backbone = 0.75):\n",
    "            * <b>Evaluate MC criterion and end Backrub MC step</b>\n",
    "        * If only Backrub moves==False (P.backbone = 0.75):\n",
    "            * Pick 1-2 residue sidechains and apply rotamers\n",
    "            * <b>Evaluate MC criterion and end Backrub MC step</b>\n",
    "<br><br>\n",
    "* Added MetropolisHastingsMover in place of GenericMonteCarloMover the protocol with Backrub and Sidechain movers referenced. Sidechain mover has prob_uniform = 1 so that random rotamers will be applied.\n",
    "* Using AddConstraintsToCurrentConformationMover with MinMover to simulate Minimize_with_cst app\n",
    "<br><br>\n",
    "* Unlike the BackrubProtocol, where P.rotamer = 0.25 and P.backbone = 0.75, the Rosettascript will always apply backrub and sidechain moves."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## April 18, 2016\n",
    "\n",
    "Add sidechain rmsd observer thing<br>\n",
    "Noah's methods in enzymology paper for backrub minization asdf<br>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## April 25, 2016\n",
    "\n",
    "Start thinking seriously about dissertation projects\n",
    "\n",
    "* Multistate enzyme design\n",
    "    * Samuel's project with the (asdf)?\n",
    "    * GEF/GAP in line with Pardeep's/Tina's stuff\n",
    "    * Mutating (asdf) to see how far in sequence identity we can go\n",
    "    * Different enzyme system (GEF/GAP) (see Lim paper)\n",
    "* Biosensors\n",
    "    * Designing new pocket for A-B dimer\n",
    "    * Design \n",
    "* Amino acid positioning\n",
    "    * Kale's enzyme thing\n",
    "    * Cripsr/Cas9 PAM sequence identification\n",
    "    \n",
    "Add a filter to remove clashing structures from runs\n",
    "Look into clashing structures to see what exactly is happening and where\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}