This page is for everyone to post their tips/hints for:

  • Wet lab

    • Protein Expression and Purification Issues (aggregation)
  • PyMol

  • Virtual screening

  • Enzyme Assay software (Graphpad Prism)

  • Liquid Nitrogen in lab

Protein Expression

Chagnes to protein expression protocols if you did not get good yields:
  1. Measure O.D.'s every 30 min to determine when the graph steepens sharply and you are in log phase
  2. Start induction with IPTG at 0.3 O.D. (this will actually save you some time) - still start the culture in the morning at 0.100 or less though.
  3. Reduce Expression (after IPTG) temperature to Room Temp or 30 deg C (need to use Biotech lab shaker for this - or shaking water bath)

Perfect Protein Buffer

Protein Information:
Protein Expression Website
Protein Expression Paper: SGC_ProteinProductionPurificationNatMethods2008.pdf

Protein Aggregation
Paper: Detection and prevention of protein aggregation before, during, and after purification. Bondos, Bicknell.

PyMol Tips:

Kaarthik's tricks:
Replace alias/placeholders in bold with appropriate commands
Many of these commands are redundant

To view the sequence of amino acids in the residue: Display -> Sequence On

To select a residue: select residue 1
To select multiple residues: select 27-30
To select and label a residue: select v12, resi 12
To select amino acids of the same type: select leu resn leu
To select waters: select wat, resn hoh
To select hydrophobic residues: select hyfob, resn leu+ile+val+phe
To select ionic residues: select ion, resn asp+glu+lys+arg
To select polar residues: select ser+asn+gln+thr
To show double bonds and aromatics: DIsplay -> Show Valences
To select beta sheets: select Bsheets, ss s
To select residues within 5 Angstroms of the active site : select active-hs, byres(mtx-hs around 5 & !HETATM & DHFR-hs) excludes all heteroatoms - ligands, substrates and waters
To select the protein excluding the active site: select pro-hs, byres(DHFR-hs & !active-hs & !HETATM &DHFR-hs)

other examples:
select ndp, resn ndp
select mtx, resn mtx
select pro, resi 1-186 selected entire protein which had 186 amino acids
select trace, resi 10+20+30+40+50+60+70+80+90+100+110 selected every 10th residue
select beta, resi 4-10+130-138+176-185
select strand1, resi 4-10
select DHFRcats, resn lys+arg selected cations
select Phe31, resi Phe31
select mtx-hs, resn mtx
select ndp-hs, resn ndp
select ndp-lc, (resn NDP & dhfr-lc) use the ampersand if the ligand of interest i.e. ndp is in more than one molecule present in the pymol visualizer.
select mtx-lc, (resn MTX & dhfr-lc)
select pro-lc, resi 1-162 & dhfr-lc
select wat-lc, (resn HOH & dhfr-lc)
select active, (byres (16dhfr-ligsX around 5))
select wat-hs, resn(HOH & 1U72)
select ndp, resn ndp
select active, byres (20ligs-Gout1 around 5 & protein_1u72)
sélect hyfo, resn val+leu+ile+met+phe+trp+try & active selects all hydrophobic amino acids around the active site

A. Right click on the atom of interest. Scrolling down to residue and selecting label will open up options to add a label to the display window. CTRL+left clicking on the label will move it a new location in the display window.
B. label 12, residue12
C. L -> residue identifier selection

set label_color, black
set label_size, 20

other examples:
label (DHFRcats & n;ca), resi+resn
label (Phe31 & n;ca), "Phe31"

Wizard -> Mutagenesis

align dhfr-lc & n. ca, DHFR-hs &
align (dhfr-lc & n;ca), (DHFR-hs n;ca)
If the above two do not work try:

split_states 16ligs-Gout2

create object x, object 1 or object 2

1. To select the active site for a specific protein and inhibitor when two or more proteins
are aligned in PyMOL:


select active1U72, (MTX1U72 around 5 & 1U72 & !HETATM)

1U72 is the protein and MTX1U72 is the of 1U72. This example selects the active site for 1U72 by putting
"& 1U72". Waters and other cofactors are not included by the command "& ! HETATM".
The "&" just means that you are adding another command. So "& 1U72" means that you just want 1U72 and not
another protein. "& ! HETATM" means that you do not want waters and other cofactors included in the active site.
This is useful when two proteins are aligned and you just want one proteins active site.

Do this command twice with its respective protein and inhibitor if you are looking for the other proteins active site.

2. For posting images to Wikispaces page from a Mac:
Screenshot Images for MacNotebook: Press Command+Shift+3
Snip Images for MacNotebook: Hold Command+Shift+4 and navigate cursor (Image will be on desktop)
--- courtesy of Ronnie S.


In order to show the coordinates of an atom - make the atom a selection and then:
$iterate_state 1, atom1, print x,y,z

To Label coordinates in PyMol:
PyMOL>from pymol import stored
PyMOL>stored.pos = []
PyMOL>iterate_state 1, (selectionname), stored.pos.append((x,y,z))
PyMOL>label selectionname, ("%5.5s, %5.5s, %5.5s") % stored.pos.pop()

OR - just open the PDB file as a text document and find the atom you want. The X,Y,Z coordinates are shown on the same line.

Selecting waters - Dr. B

$select wat, resn HOH

Selecting everything that is not protein - Dr. B

$select hetero, resn HETATM

Selecting protein but not including active site nor other atoms (heteroatoms) so that you can display the rest of the protein as cartoon and transparent - Dr. B

(this assumes your PDB protein is named '1U72' and that your active site selection is 'activesite'. Modify accordingly if yours are different.
$select pro, (1U72 & !activesite &!HETATM)


remove not (alt ''+A)
alter all, alt=''

Virtual Screening Tips:

Converting third-party compound IDs to ChemBridge IDs

Logging in to the DDFE

  1. It is best to login on campus. It is also usually works best when you are in WELCH where the DDFE actually is. But Painter should do well. However, I don't know about other buildings on campus.
  2. It is best to use a 'hard-wired' connection (i.e. an ethernet cable). Wireless does sometimes work - but it can be really slow.
  3. Use Virtual Private Network (VPN) in order to connect. This should work with both wired and wireless connections.
  • It requires your UTEID. There is the browser version - but you may have to 'install' it on you computer to make it work.



XQuartz for visual X11 window on Macs
download the cisco client.

I fixed the problem to log onto the x server from a mac with the new OSX version (after Mountain Lion Update)
. attached is the link for the download site in case other mac users want to use x11 - Ling
New Link:


FUGU - eat it and die!
Download Fugu to have a program that will give you a graphical folder interface like SSH
use Terminal (part of the OS) to do command line interface. Once you open terminal, type this to get a connection (text only):

Use X11 (part of the OS) to get a graphical X window.
Type this to get a connection (with graphical capability - i.e. for GOLD GUI):
$ssh -X

MacFusion ?? also

Quick, duck it's a.......... CYBERDUCK

This is mainly for those of you that utilize a "Mac-in-trash" (- Dr. B). You can technically do virtual screening from home on a Mac! This is what I did a lot my freshman year, because I could never finish by 9 pm, and I wanted to work from home. You are going to need to download this program called Cyberduck. You can find that here: You also need to make use of the Terminal application on your Mac (Launchpad --> Other --> Terminal). In Cyberduck you are going to click Open Connection. Change FTP to SFTP. Enter for the server. Change the port to 22. For user name and password use your UTeid and password that you used in lab. Cyberduck acts like WinSCP in that you can drop and drag files to your desktop. Using Cyberduck you can navigate to the folder with the reports needed for your lab report. The Terminal acts as the command prompt where you can enter in prompts like sh etc (putty). In the Terminal application type in: ssh (except you're going to be using your UTeid). Enter in your account password and you should be golden. If you have any questions ask Katherine or Brandy, because they are experts. Or you can just email me. There are better methods and programs, but this is what I use, and I thought it would be helpful.

Michael T.


  • Apple-Shift-3 takes a screenshot of the whole screen. Another one that most people don't know about is Apple-Shift-4, which produces crosshairs that can be dragged to grab just a portion of the screen.
Then look at your desktop - there should be a new file there that is your image.

OS X also comes with the Grab utility, with which you can grab a selection, a window, the screen, or a timed screenshot.

  • Apple-Shift-4 then spacebar takes a screenshot of just a particular window. And Apple-Shift-4 then Control copies the screenshot to the clipboard.

VIRTUAL SCREENING ON YOUR MacIntrash (courtesy of Jairo V. 10/21/2014)

Also, after spending hours working on a Mac and following the unclear instructions found in wikispaces, I found an easier way to run virtual screening jobs on Mac.
  • Download VPN (Virtual Private Network) The website is
  • Alternatively, you can download the "Cisco AnyConnect Secure Mobility (VPN) Client" found at the BevoWare website under MacOS Utilities and install the program.
  • Signing into the Cisco VPN Client:
    1. Open up Cisco VPN Client Application.
    2. Type in “” in the “Connected To” section.
    3. When prompted, type in your username (UTeid) and your password (normal UT password).
    4. You are now connected, whenever you’re finished, just click Disconnect
Note: I am not sure if I was supposed to sign into the “” server using the VPN Client or not? - NO

Note: Download the program from the website not from the App Store because its free compared to $23.
  1. Install and open Cyberduck
  2. To sign in:
i. Click “Open Connection”, a window with options should come up.
ii. From the first drop down menu change the “FTP (File Transfer Protocol)” to “SFTP (SSH File Transfer Protocol)”
iii. Server:
    • Port: 22
    • Username: UTeid
    • Password: (UTeid with uppercase letter and shifted last number)
iv. Click Connect
  • Editing scripts can be done by selecting the script and clicking “Edit” on the top menu bar, once finished just click save and exit out of the script.
  • I couldn’t figure out how to transfer files from folder to folder, so I ended up downloading the files from the source and dropping them to their destination and that worked

  • Terminal will be used to run the gold jobs.
  1. Open up Terminal, use spotlight and search terminal.
    1. Commands:
i. ssh into server) (use same username and password as Cyberduck)
      • (don’t include the #, just your UTeid)
ii. ls (list of files in current folder)
iii. cd ./#nameoffolder (opens a file or folder in current folder) (don’t include the #)
iv. sh (runs gold)
v. ps (status of the job)

FileZilla - for DDFE access

ACA 1.126 has a computer lab that you can do virtual screening at. It has Xming, XLaunch, and Putty - but no WinSCP
So, you need to go to Bevoware on the Utexas site and download it and INSTALL IT TO THE DESKTOP!. THen you will have a file sharing utility.

IF, your GOLD jobs don't work......

since you are probably using a Windows computer to edit your , the file may not be readable to the LINUX based operating system that is running the SGE (Sun Grid Engine) to submit jobs. So, you need to convert your scripts to UNIX format (don't worry - it is easy). After you have made the changes to your script file and are ready to run your GOLD job, just type:


You should do this for each script file you want to run. I think you won't have to have multiple script files for each job anymore with this change!

$chmod u+x

This give you permission to run your script.

Concatenating Your Bestranking.lst files:

If you would like to use a script to concatenate your Bestranking files instead of having to manually put them together, you can use Christina's script in the /home/chem204/scripts directory.
Open up the script and look at it first to read how to execute it.

Monitoring Your Jobs

To check if your gold job is running, type:

To show all users's Jobs.

This does work though. "*" is the WILDCARD character and means 'all users'

$qstat -u "*"

Dr. B

To see if your specific job is running, type:
$qstat -u userID

To kill your job, type:
$qdel jobID
  1. For example, jobID = 2708
$qdel 2708

How to break up your table into columns when using EXCEL for Bestranking.lst

When using Excel to examine your ligands
Save Bestranking.lst to desktop
Open EXCEL fresh (i.e. without a file yet)
Then go Open your Bestranking.lst that is on the desktop
The Text Import Wizard will open - use 'space' or 'tab' delimited to see if it will break each score category into individual columns.
Then sort your ligands
Make a new column for numbering, then number your ligands (e.g. #1 .........1,000)
Adjust the Column Headers so that they read properly

Finding Homologous Structures of your protein in the PDB
To find homologous structures - go into the PDB and do an Advanced Search. One of the options on the pull down window on the Advanced Search page is Enzyme Classification. Type in your EC number.
It will give you a list - then narrow that down on the left hand side by other bacterial/protozoan organisms (pick ones that are more taxonomically similar to your organism)

Active Site Definition for Virtual Screening

Defining the Active Site by creating an Object in PyMol. a.k.a. - the 'dummy ligand' approach
- (courtesy of Sadhana B.)

NOTE: I do not know how to do this on a MacIntrash. I will give you a dollar if you figure it out and add the instructions here. - Dr. B

This is useful if you do not have a ligand in your active site and need to make a ligand to define your active site for ICM or GOLD Docking. It is difficult to define a point in the middle of a space where there are not atoms, so you can create the 'dummy ligand' as a new object and then move it to the center of your active site.
  • Also good for a Homology Model that does not have a template with a ligand in it.
  • Also, use this if your actual Homology Model only has a metal in it but no ligand. You can't use the metal as the center of the site if you are going to keep the metal in the docking runs, so you need to create a 'dummy ligand' nearby.
  1. Write this down in your lab notebook (you'll forget what you did)
  2. Go to Pymol
  3. Pull up protein (can go to command line and type $ fetch PDB identifier) i.e. $ fetch 2cm1
  4. Find where the active site is (usually where metal ions are in a metallo-enzyme) - but you may have to view homologous structures from other organisms to figure out where the active site is.
  5. Go to Build >> Fragment >> Choose object, preferably something cyclic that is easy to see (i.e. cyclopentadiene)
  6. To drag object to the active site, go to 'A' tab for the selection >> Drag Coordinates. (click on 'Indicate' to highlight it)
  7. Then hold the Shift key and use the middle mouse key to move the object. And Shift key + left mouse key to rotate
  8. Once you are satisfied that the location of your object is in the center of the active site, go to file >> save molecule. Select your new object and the protein to save as PDB (use 'Ctrl' to click more than one item in the list)
  9. You can now copy this new protein file into your virtual screening folder to do docking wherein you define the active site with an existing ligand (i.e. the one you just made).

  • Alternatively, if you want to get the X,Y,Z coordinates of one of the atoms in your 'dummy ligand' to use in GOLD docking for binding site definition, just change your Selecting Mode to 'ATOMS' in PyMol, click on the atom and then look at the text viewer to see what the atom name is. If that is not clear - then use the 'L' tab for that selection to label the atom using Atom Identifiers >> ID (using the other options 'rank' or 'index' may be incorrect).
    • Then in a TEXT EDITOR (i.e. Wordpad) open the PDB that you saves in the STEP above and search for the atom and its X,Y,Z coordinates.
    • When you set up you GOLD job - use the X, Y, Z coordinate you just got.

Defining the Active Site of Homology Model for GOLD using the ligand already present in your Template pdb
-------(NOTE: can only do this if your Template has a ligand in it!). If no ligand - use the method above (Creating an dummy ligand))
  1. Write this down in your lab notebook (you'll forget what you did)
  2. Open Template and Homology model in the same PyMol session (these need to be .pdb files not .pse files)
  3. Align your Template to the Homology Model (not the other way around. You don't want the coordinates of your Homology Model to be messed up for docking - DON'T MOVE THE HOMOLOGY MODEL!) --------- use the 'A' tab, 'Align'
  4. File >> Save Molecule >> One Molecule -- You only need to select the ligand to save. --------- save as ligand_aligned.pdb (or something memorable) and copy to your GDocs.
  5. Find an atom in the ligand that you can use as the center of your binding site (use your judgement here to find an atom in the center). Change your Selecting Mode to 'ATOMS' in PyMol, click on the atom and then look at the text viewer to see what the atom name is. If that is not clear - then use the 'L' tab for that selection to label the atom using Atom Identifiers >> ID (using the other options 'rank' or 'index' may be incorrect).
  6. Open the newly made .pdb file for the ligand in Wordpad.
  7. Get the X, Y, Z coordinates of this atom from the Wordpad file and then write them down
  8. When you set up you GOLD job - use the X, Y, Z coordinate you just got.

App for Molecular Viewing on your 'smart' phone

- From the company that makes ICM

MolSoft has released iMolview ( iMolview is an App that enables you to view protein, DNA, chemicals, and molecular documents on mobile devices in 3D. iMolview is free and can be downloaded here:

How to create a log sheet with all your activity on Linux

Tired of having to worry about time stamps? Forgot what command or shell script you used and when? Using this tip you can have linux create a text file containing all of your activity with timestamps for each shell session. Type the following into your command line:

script filename.txt

This will start a new shell session where everything is recorded in a file called "filename.txt", you can choose a different name but make sure to include .txt at the end. Next to add the timestamps you can modify the prompt variable PS1 by typing the following (the spacing in this is important):

PS1="\@ \h \$ "

Now you can use the shell like normal and all of the output on your terminal will be copied into a file called "filename.txt". When you are done make sure to type:


This will complete the file! Now you have a file called "filename.txt" in your working directory with a copy of your whole session with timestamps. If you decide you want to start another shell session and add things to your "filename.txt" file instead of starting a new one then type the following:

script -a filename.txt

Look up the man page for script for more info on interesting options!

Lightshot - Screenshot Tool for Windows and Mac

This is a very lightweight program that you can install on your computer to take screenshots. If you have it installed all you have to do to take a screenshot is press the "print screen" button on your keyboard and then you can select and manipulate the area on your screen you want to take a screenshot of. With this tool you can save, copy, print, do a reverse image search, or add arrows, text, or shapes to your selected area. Easier to use and more powerful than snipping tool!
- Luis

Graphpad prism
- is on the old Robertus laptop in lab
username: VDSclass
passwerd: same as other laptops

login info (remote desktop using Windows - not sure if it will work from a Mac or Linux)
-- find remote desktop in Windows start menu from your computer and enter this info.
username: vdsstationPC\vdsclass
passwerd: same as to get into laptops in lab

10 Tips for Safely Using Liquid Nitrogen in the Lab