Session 5

 

 

 

Presentations

disorder_part1

disorder_part2

PART 1

Prediction of protein disorder

Exercise 1

DISPROT database and analyzing calcineurin A

You can start from UNIPROT, it has a link to DISPROT
or search DISPROT directly by keyword, or sequence.

  • What kind of information can you find on this page?
  • Which regions are marked as disordered in DISPROT?
  • Which regions are marked as ordered DISPROT?
  • By what experimental techniques?
  • What is the role of the disordered segments?

 

Exercise 2

Go to the search page in Disprot
and select source organism

Can you find proteins in Disprot from
–       viruses
–       Mycobacterium
–       Human
–       Plasmodium
–       your favourite species?
Look at some of the examples

Exercise 3

Disorder prediction methods

Collect prediction outputs for calcineurin A using

– IUPred http://iupred.enzim.hu
– Globplot http://globplot.embl.de/
– PONDR-FIT http://www.disprot.org/pondr-fit.php
– PredictProtein http://ppopen.informatik.tu-muenchen.de/
(MD, UCON, Norsp, Profbval)

  • and any other method you like.

The input can be:

  • the amino acid sequence in FASTA format
  • amino acid sequence in raw format (without header)
  • UNIPROT ID or accession number

Please note, some methods are sensitive to line breaks !
Minimum and maximum length of sequence !

NOTE: Cutoff values, False positive rates, different types of output

  • Do the predictions agree with the experimental characterization of disorder?
  • Do the predictions agree with each other?

 

Exercise 4

DISPROT DP00039 : a highly disordered protein

  • Predict protein disorder for DISPROT DP00039
  • Count number of positively charged amino acids
  • Count number of negatively charged amino acids
  • Calculate net charge
    (or use the protparam server)
  • Check low complexity segments (you can take these from PFAM through uniprot)
  • Check PFAM domains

Is there a contradiction between PFAM domain assignments and predicted disorder?

Exercise 5

One of the main applications of disorder prediction methods is to find
suitable targets for structure determination.

  • Which region of this protein you would try to crytallize?
    P77173
  • Check the Uniprot entry, which region could be be solved?

 

Exercise 6

Check Pfam domains and disordered regions in these three proteins

  • Q8IJX0
  • Q54TH9
  • Q07960

What type of domains do they have in common?
Where are the disordered regions located, if there are any?
What is the likely relationship among these proteins?

 

 

PART 2

Exercise 1

Disordered binding regions in the N-terminal region of p53

The N-terminal region of p53 contains at least three binding regions:

MDM2: 17-27
RPA70N: 33-56
RNAPII: 45-58

  • Use ANCHOR http://anchor.enzim.hu
  • run ANCHOR on p53
  • Are the experimentally determined disordered binding regions located in the N-terminal agree with the known disordered binding regions?

You can add motifs to ANCHOR prediction

For example search for the MDM2 motif and check if it coincides with disordered binding region

F…W..[LIV]

 

Exercise 2

The nucleoprotein of measles virus

http://www.uniprot.org/uniprot/Q89933

Which region is ordered, which region is disordered?
Does this protein have disordered binding region?

(use ANCHOR)

 

Exercise 3

Find beta catenin in Uniprot (CTNB1_HUMAN)

  • Go to the natural variants section. You will see multiple sites are associated with cancer. Where are most of them located?
  • Characterize the site of mutations structurally and functionally.
  • Use the ANCHOR,
    SlimPred and
    Morfpred servers

Exercise 4

Identification of binding sites based on patterns of conservation