My software notes

October 19, 2011

NMR protocol (recipe)–make stretch gel for protein RDC measurement

Filed under: recipe_protocol — kpwu @ 7:01 pm
Tags: , ,

I bought the starter kit (NE-373-B-6/4.2) at NewEraNMR for RDC experiment on Varian NMR (with 700 MHz grade) and I decided to stretch the gel a lot, so 6mm-wide gel chamber and “6mm-4.2mm gel funnel” are included.

The NMR tube seems to be more fragile than Wilmad products I often used, so I also ordered additional 5 tubes (both ends are opened).

Since the special point of the NMR tubes used for this stretch gel is “two-end opened”, one can find glass factory to cut the regular NMR then no need to buy this special tube.

To users who can use both Varian and Bruker NMR machines:  you just need to buy some “End Gel Plug” for Bruker machine. It’s not necessary to buy two starter kits.

The picture shown left is an example I made and the bubble near the bottom of gel showed up after 3 days. User may have to to worry the time-dependent changes of the stretched gel.


Required solution:

  • 30% acrylamide (w/v) in distilled H2O
  • 0.8% bis-acrylamide (w/v) in distilled H2O
  • 10% APS (w/v), APS = ammonium persulfate
  • TEMED as catalyst
  • Commercial kit to make stretch gel in NMR tube (page 16 in NewEraNMR’s catalog)
Recipe for 1ml mixture: (ul used at here means micro-liter)
  • 6% gel:
    • 30% acrylamide: 200 ul
    • 1% bis-acrylamide: 80 ul
    • 10% APS: 10 ul
    • TEMED: 0.1 ul
    • H2O: 710
  • 4.2 % gel:
    • 30% acrylamide: 140 ul
    • 1% bis-acrylamide: 56 ul
    • 10% APS: 8 ul
    • TEMED: 0.1 ul
    • H2O: 796
You may note that I made the ratios of acrylamide:bis-acrylamide:APS constant between 6% and 4.2% gel. Just try to avoid too much “unreacted” APS and acrylamide inside the gel.
Protocol of gel preparation:
  1. Mix the above recipe and load 800 ul into the gel chamber, wait for 2 hour for gel formation. The “un-used” 200 ul mixture is used to check if the gel is formed.
  2. Gently remove the gel from chamber using the extrude rod provided in the starter kit and use cleaned knife to cut the gel to a proper length. I made both 1.5 and 2mm gels to next step.
  3. Wash the short gels with 15 ml distilled water for 1 hour and repeat 3-5 times. This step is to try to remove unreacted acrylamide, TEMED, and APS inside the gel.
  4. Wash the gel with 15 ml NMR experimental buffer for 1 hour, repeat 3 times. (or do more volume/times and skip step 3)
  5. Soak the gel with 600 ul NMR buffer which has 10% D2O inside and leave it overnight.
  6. Remove the buffer and  transfer~300 ul concentrated NMR sample (10% D2O included) to the tube containing short gel.Wait for free diffusion for overnight.
    I usually use ~1 mM NMR sample and the final protein concentration left in the solution is less than 600 uM suggesting 300-400 uM protein are inside the gel.
  7. Assemble the gel chamber with funnel, load the short gel (with protein and 10% D2O) into it. Follow the guide in the NewEra’s catalogue, it should be easy to move the gel from chamber to NMR tube.
  8. Once the gel is fully inserted into the NMR tube, it’s stretched and quickly insert the “End Gel Plug” then insert the assembled “rod-end up plug” to keep the gel at proper position.

As my personal experiences, 6% gel and 4.2% gel returned me 2H splitting constants of 6.2 Hz and 2.5Hz, respectively.


Bruker’s Topspin running on Mac

Filed under: mac,softwares and scripts — kpwu @ 2:21 pm
Tags: , ,

Balance the report as I just made a post for Agilent’s VnmrJ. 🙂

I found from a web blog article that introduces the news–Bruker’s Topspin 3.1 for Mac.

Bruker’s brief introduction of Topspin for Mac machine includes features:
(note on 10/29/2012, the introduction seems gone at Bruker’s website)

  • Mac OS X platform for TopSpin processing
  • Inspire a new generation of NMR users, experts and students alike
  • Entirely programmed in the native Apple Mac OS X environmentData analysis, processing and simulation features for efficient small molecule and structural biology research
  • Ideal data processing and analysis aid for NMR training

Snapshots of Topspin running on Mac linked to Bruker’s example images show at here:

(note on 10/29/2012 the image was removed from the blog)

According to Bruker’s purchase link, the price is $99 of academic license for one Mac machine. The computer MAC address code must be provided meaning one license is only available for one Mac machine!!

The current version (Oct of 2011) Topspin for Mac required “Core 2 Duo” type CPU and OS 10.6 (snow leopard and later). My old macbook pro with “core duo” is not available this time. Sad. I can only try to install it on my iMac.

Maybe I should email Bruker to request something like 30-60 days free trial and make some blog articles for “Bruker Topspin on Mac” as an exchange.

October 18, 2011

Agilent’s VnmrJ 3.2 implements NMRPipe

Filed under: NMRPipe and NMRview — kpwu @ 5:06 pm
Tags: , , ,

As just announced at the Agilent’s web blog “SPINSIGHTS”, the coming VnmrJ 3.2 will provide complete implementation of NMRPipe. This will be very useful for most of biomolecular NMR users.  I always run NMRPipe on the console linux to preview the 2D HSQC or some 3D data and direct compare to some data I already processed by NMRPipe for conveniency. Checking data in VnmrJ with 2-column view is not fun, honestly.

The blog article “Advanced Data Processing—Implementing NMRPipe Within VnmrJ 3.2” explains the details and also provide example figures.

Some images shown below are linked to the figures posted at the SPINSIGHTS.

October 11, 2011

[web server] protein structural analysis

Filed under: servers — kpwu @ 7:00 pm
Tags: ,

PDBsum provides service that users can upload their calculated structures for analysis including ramachandran plot, sequence+secondary structural element, etc. The figure below is an example. For self-generated structures, users need to click the link “generate” to upload your PDB files. To know the structural information of published biomolecules at Protein Data Bank, just type the pdb access to get the details.

After a short waiting (to this example, i just waited 10 minutes to get analysis done), an email sent from PDBsum provide users to look the details of the submitted structure. I clicked link at the left down corner to see the information of sequence and location of secondary structures. Other analysis including topology, SAS search and rama plot are also available for self-uploaded structures.

Another web server called “POLYVIEW 2D” also provides similar service. Users can look at the structural information of published structures deposited at PDB or upload their own structures (in PDB format) to draw either single or ensemble views. Two snapshots are shown below.

NOTE. POLYVIEW-3D, another service provided by the same team is an useful web server to generate pictures of target biomolecules without opening the molecular viewers on your own computers. It is very convenient for me when I am away from lab and wanna to see some snapshots of proteins. Highly recommended.

October 6, 2011

[PyMOL] assign secondary structural regions

Filed under: Uncategorized — kpwu @ 3:25 pm
Tags: , , ,


Sometimes I displayed the cartoon view of a protein in PyMOL as the figures shown below that I don’t really see the helix and strands.

It often happens when I merged 10 or 20 structures into a NMR ensemble file. All 20 structures are aligned but the secondary structural regions are not clearly displayed in PyMOL (commands: set all_states, on)

PyMOL provides two simple means to quickly draw helices and strands on the target proteins.

1. use command “dss”.

Simply type “dss” in the command line, PyMOL automatically calculates, rebuilds and draw the secondary structural regions. If the “all_states” is turned on, all structures will also be displayed properly. Here are two figures show the results of “dss” with single and all 20 states of my target protein.

The command “dss” provides simple but quick way to draw helical and strand regions, however, using other programs such as “DSSP” or “STRIDE” to clearly define the regions of helices and regions is highly recommended.

2. Use command “alter”

As I just said, “dss” doesn’t work perfectly. In this example case, I found length of first two sheets are not long enough, and the loop is a bit longer. I can use command “alter” to manually assign the secondary structural regions of my target proteins.

alter 3-9/, ss=’H’
alter 12-18/, ss=’H’

The above 3 lines are the commands I typed in and the sheets are now longer (see the right up corner). To change the length of loop and sheets, just type alter xx-yy/, ss=’L’, and alter aa-bb/, ss=’S’, respectively.


October 5, 2011

[PyMOL] customized display of helices and strands

Filed under: pymol/ molmol — kpwu @ 11:43 am
Tags: , ,

The  default view of cartoon presentation in PyMOL is provided below.

PyMOL provides commands “cartoon_oval_length“, “cartoon_oval_width“, “cartoon_rect_length” and “cartoon_rect_width” to customize the display of helices and strands.

PyMOL users should be aware that the 4 commands can’t be combined with “cartoon_fancy_helices” and “cartoon_fancy_strands”.  Once the fancy mode is turned on, the manually customized width/length will be changed back to default number.

The first few examples will be related to the oval view

command: set cartoon_oval_length 0.2

command: set cartoon_oval_length 1

command: set cartoon_oval_length 1.7

We can also change the thickness of the helix in our target by using “cartoon_oval_width”

The figure below shows the default thickness of helix.

command: set cartoon_oval_width 0.8

command: set cartoon_oval_width 1.5

Now. let me show you some examples to change the width and thickness of strand by using PyMOL.

command: set cartoon_rect_length 0.3

command: set cartoon_rect_length 0.7

command: set cartoon_rect_length 2

command: set cartoon_rect_width 0.1

command: set cartoon_rect_width 0.9

command: set cartoon_rect_width 1.6

October 4, 2011

Some cartoon views of proteins shown in PyMOL

Filed under: pymol/ molmol — kpwu @ 6:56 pm
Tags: ,

Example of cartoon modes of protein shown by PyMOL

Example protein: 1UA8, LolA

By default, the cartoon mode in PyMOL shows protein structure like this way:

We can change the background to white by typing “bg_color white” in the command terminal and also change the color of protein based on types of secondary structures in the color options (the “C” button on the right side of this object).


The cartoon view of this protein shows red, yellow and green colors for helical , strand and coil regions respectively. However, the interior and exterior sides of helices and sheets have colors which doesn’t provide clear view. So one way in PyMOL to create the difference is called “highlight color”. I typed “set cartoon_highlight_color lightorange” to change the color of internal side of helices and side of strands.




When rotating the protein, the zoom-in view of helices shows the border of each helix is smooth. There are commands called “cartoon_fancy_helices” and ” cartoon_fancy_sheets” that provide fancy borders of the helices and strands.


By typing “set cartoon_fancy_helices, 1” and “set cartoon_fancy_sheets, 1” (1 is on, and 0 is off in PyMOL, always). The  cartoon of helices has different views:



Another way to present helices is the old fashion cylindrical/rod view by typing “set cartoon_cylindrical_helices, 1“.

One more example to show is the “discrete color”! You may find that residues in the loop regions sometimes colored in yellow or red (not green) when connected to strands or helices. The mode of discrete color is turned off at this moment. The following two figures present activated and inactivated mode of “discrete color”. Personally, I like to turn on the discrete color mode all the time.

ON mode (set cartoon_discrete_colors, 1)

OFF mode (set cartoon_discrete_colors, 0)

Blog at