Local or global alignment?

[Alessandro Maiorana]

Hi all,

I have a little problem. I have a big protein (apoB-100) sequence
(about 4500 aa) and I would like to find in this protein a binding
site of hormone. Doesn't exist the structure of it but I had obtained
a low resolution model (SAXS technique), and this model showed that an
area of protein is affected by hormone. Unfortunately this model is
not to high resolution and I can not identify the sequence or domain
affected by it. The my idea is to align the receptor sequence that
bind this hormone (about 250 aa), of that there's the crystallography
structure, and 1) the complete sequence of apoB-100.
I think a local algorithm to start. Do you have any hint?

2) I will align only the specific domain of apoB-100 (about 500 aa),
that, in my experiment, is affected by the presence of hormone. I
think to use both algorithms of alignment (local and global). Do you
have any hint?

I have already tried to use the on-line program EMBOSS and I had
contradictory results. In your opinion, Is there a rapid way to obtain
a good alignment? How can I stabilize if an alignment is good or bad
(score or similarity or identity or both)? Best regards

A.M.

[akhilesh Datt pandey]

Dear Alessandro,

You can use Mega software to align your sequence there are much more softwares like clustal w clustal x everyone having some different criteria ... with Mega you can do this easily.

But it would be very difficult to predict that which hormone is going to bind with your protein ..it needs a lot of logical analysis by using more than 10-15 online databases and their software programs. 

You should start looking various aspect of your protein http://workbench.sdsc.edu/ you can also use this for different types of alignment and little bit other sort of analysis.

Sincerely 

Akhilesh

  

 

--
You received this message because you are subscribed to the Google
Groups "PredictProtein-Science" group.
To post to this group, send email to
predictprot...@googlegroups.com
To unsubscribe from this group, send email to
predictprotein-sc...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/predictprotein-science?hl=en

This group is affiliated with the PredictProtein server - http://www.predictprotein.org/.
PredictProtein is an Internet service for sequence analysis and the prediction of protein structure and function from sequence.

[vogelhugel]

dear,

thx for your answer and i want to know hou to enter the program

Biology Workbench 3.2.

when i click the link it said i should have the password.

thank you!

yours sincerely

 

2010-11-21

---

vogelhugel

---

发件人： akhilesh Datt pandey

发送时间： 2010-11-21  14:59:26

收件人： predictprotein-science

抄送：

主题： Re: [predictprotein-science:433] Local or global alignment?

To unsubscribe from this group, visit
http://groups.google.com/group/predictprotein-science/subscribe?hl=en_US

For more options, visit this group at
http://groups.google.com/group/predictprotein-science?hl=en
 
This group is affiliated with the PredictProtein server - http://www.predictprotein.org/.
PredictProtein is an Internet service for sequence analysis and the prediction of protein structure and function from sequence.

To get PredictProtein news and updates follow us on twitter:
http://twitter.com/#!/rostlab

[SD]

Hi vogelhugel,

 

There is link to register yourself.

 

http://seqtool.sdsc.edu/

 

 

Clck on register and follw the instruction.

 

 

Hope that’s sorts out your issue.

 

Regards,

P.Sundeep Kumar

[akhilesh Datt pandey]

You need to register in the biology work bench ...and you can create
an account and you can save your data also.

look your work is critical ... Myself i predicted 7 protein-protein
intraction by bioinformatic analysis ...with lot of hell logics ..and
than i proved all those intraction by using wetlab method perticularly
co-immunoprecipitation..

so i know it is very typical type of work.

sincerely
Akhilesh

On 11/20/10, vogelhugel <vogel...@126.com> wrote:
> dear,
> thx for your answer and i want to know hou to enter the program
> Biology Workbench 3.2.
> when i click the link it said i should have the password.
> thank you!
> yours sincerely
>
> 2010-11-21
>
>
>

> vogelhugel

[vogelhugel]

many thx!

 

 

2010-11-22

---

vogelhugel

---

发件人： akhilesh Datt pandey

发送时间： 2010-11-22  04:14:58

收件人： predictprotein-science

抄送：

主题： Re: Re: [predictprotein-science:436] Local or global alignment?

[shankar chatterjee]

If your sequence has the signature sequence it should have been identified at NCBI/EMBL site and your send line of defense appears to be OK
thanks

[varun bhaskar]

Hi A.M, 

              Just read your problem. I suggest you use a software called Pymol which is available for free for academic use. It is a structure viewer like spdbv. You can then load the receptor structure in this program and then align the ApoB-100 sequence on the structure. It is very easy process and what the program does is fit the ApoB-100 sequence onto the receptor and you can literally edit the sequence and automatically see the change in fit. You can calculate RMSD values between the sequence and the receptor structure to confirm the quality of the fit. 

Hope it helps!

Varun

To unsubscribe from this group, visit
http://groups.google.com/group/predictprotein-science/subscribe?hl=en_US

For more options, visit this group at
http://groups.google.com/group/predictprotein-science?hl=en
 
This group is affiliated with the PredictProtein server - http://www.predictprotein.org/.
PredictProtein is an Internet service for sequence analysis and the prediction of protein structure and function from sequence.

[Alessandro Maiorana]

Hi, Varun

alas, I don't believe Pymol is free...I would like to try Pymol but I'm not able to upload the academic version. In your opinion, I should align the tertiary structures if I have well undertood but I don't know ApoB-100 structure. Could you explain me better? Could you suggest an alternative software to Pymol? Thanks,

Best Regards

Alessandro Maiorana

[varun bhaskar]

Hi Alessandro, 

Check this link http://pymol.org/educational/. It is the Pymol version for academic use... follow the instructions and u can get a copy of the software. As your saying that you have a crystallographic structure of the receptor of the harmone, I suggest you use that in pymol and then align that known structure with the binding region of the hormone derived from ApoB-100. As you already have a low resolution structure, you at least have an idea of the tertiary structure which will help you roughly to improve your alignment.. ! It is a lot of work based on judgment and very specific to the molecule in question. 

 spdbv (http://spdbv.vital-it.ch/) could be another software that you can use..! they arent a 100% accurate but they are always the first step in making rough models before you can do energy minimization and all that to get a very good, high resolution model ! 

If you are really having a tough time, we could work out a time to maybe have a skype conversation or something. I am quite happy to help.

Regards

Varun