CHAPTER 31
Prediction of Tertiary Structure of Protein: Ab Initio Approach

CS Mukhopadhyay and RK Choudhary

School of Animal Biotechnology, GADVASU, Ludhiana

31.1 INTRODUCTION

Protein folding is dictated by the physical forces acting on the atoms of the protein. In general, the most accurate way of formulating the protein‐folding or structure prediction problem is in terms of an all‐atom model subject to the physical forces. Energy functions compatible with the protein representation are considered during the ab initio approach. Faster algorithms are developed to search the best‐fitting formation, in order to minimize the energy function while predicting the tertiary structure ab initio.

31.1.1 Principle

To predict the protein structure based on physical principles (rather than comparative homology using the previously reported structures). The structural conformations that minimize the energy function are evaluated for the structures that the protein is likely to adopt under native conditions.

31.2 OBJECTIVE

To predict the tertiary structure of a peptide using an ab initio approach with the online tool RaptorX.

31.3 PROCEDURE (RAPTORX)

  1. Suppose we need to predict the structure of the given protein sequence (Bubaline Dicer I; NCBI Protein Accession Number BAP00765.1): >BAP00765|Bbu_Dicer1

    M K S P A L Q P L S M A G L Q L M T P A S S P M G P F F G L P W Q Q E A I H D N I Y T P R K Y Q V E L L E A A L D H N T I V C L N T G S G K T F I A V L L T K E L S Y Q I R G D F N R N G K R T V F L V N S A N Q V A Q Q V S A V R T H S D L K V G E Y S N L E V S A S W T K E K W N L E F T K H Q V L V M T C Y V A L N V L K N G Y L S L S D I N L L V F D E C H L A I L D H P Y R E I M K L C E N C P S C P R I L G L T A S I L N G K C D P E E L E E K I Q K L E K I L K S N A E T A T D L V V L D R Y T S Q P C E I V V D C G P F T D R S G L Y E R L L M E L E E A L N F I N D C N I S V H S K E R D S T L I S K Q I L S D C R A V L V V L G P W C A D K V A G M M V R E L Q K H I K H E Q E E L H R K F L L F T D T F L R K I H A L C E E H F S P A S L D L K F V T P K V I K L L E I L R K Y K P Y E R Q Q F E S V E W Y N N R N Q D N Y V S W S D S E D D E E D E E I E E K E K P E T N F P S P F T N I L C G I I F V E R R Y T A V V L N R L I K E A G K Q D P E L A Y I S S N F I T G H G I G K N Q P R N K Q M E A E F R K Q E E V L R K F R A H E T N L L I A T S I V E E G V D I P K C N L V V R F D L P T E Y R S Y V Q S K G R A R A P I S N Y V M L A D T D K I K S F E E D L K T Y K A I E K I L R N K C S K S V D T G E A D T E P V V D D D D V F P P Y V L R P E D G P R V T I N T A I G H V N R Y C A R L P S D P F T H L A P K C R T R E L P D G T F Y S T L Y L P I N S P L R A S I V G P P M S C I R L A E R V V A L I C C E K L H K I G E L D D H L M P V G K E T V K Y E E E L D L H D E E E T S V P G R P G S T K R R Q C Y P K A I P E C L R E S Y P R P G Q P C Y L Y V I G M V L T T P L P D E L N F R R R K L Y P P E D T T R C F G I L T A K P I P Q I P H F P V Y T R S G E V T I S I E L K K S G F T L S L Q M L E L I T R L H Q Y I F S H I L R L E K P A L E F K P T D A D S A Y C V L P L N V V N D S S T L D I D F K F M E D I E K S E A R I G I P S T K Y S K E T P F V F K L E D Y Q D A V I I P R Y R N F D Q P H R F Y V A D V Y T D L T P L S K F P S P E Y E T F A E Y Y K T K Y N L D L T N L N Q P L L D V D H T S S R L N L L T P R H L N Q K G K A L P L S S A E K R K A K W E S L Q N K Q I L V P E L C A I H P I P A S L W R K A V C L P S I L Y R L H C L L T A E E L R A Q T A S D A G V G V R S L P V D F R Y P N L D F G W K K S I D S K S F I S I A N S S S A E N E N Y C K H S T I V V P E N A A H Q G A N R T S P L E N H D Q M S V N C R T L F S E S P G K L Q I E V S T D L T A I N G L S Y N K S L A N G S Y D L A N R D F C Q G N H L N Y Y K Q E I P V Q P T T S Y P I Q N L Y N Y E N Q P K P S D E C T L L S N K Y L D G N A D T S T S D G S P V T A A V P G T T E T G E A P P D R T A S E Q S P S P G Y S S R T L G P N P G L I L Q A L T L S N A S D G F N L E R L E M L G D S F L K H A I T T Y L F C T Y P D A H E G R L S Y M R S K K V S N C N L Y R L G K K K G L P S R M V V S I F D P P V N W L P P G Y V V N Q D K S N T E K W E K D E M T K D C M L A N G K L D D D F E E E E E E E E D L M W R A P K E D A D D E D D F L E Y D Q E H I K F I D N M L M G S G A F V K K I S L S P F S A T D S A Y E W K M P K K S S L G S L P F S S D F E D F D Y S S W D A M C Y L D P S K A V E E D D F V V G F W N P S E E N C G V D T G K Q S I S Y D L H T E Q C I A D K S I A D C V E A L L G C Y L T S C G E R A A Q L F L C S L G L K V L P V I K R T D R E K A M C P T R E N F T S Q Q K N L S G S R A A A S G A G Y R A S V L K D L E Y G C L K I P P R C M F D H P D A D R T L R H L I S G F E N F E K K I N Y R F K N K A Y L L Q A F T H A S Y H Y N T I T D C Y Q R L E F L G E P I M D Y L I T K H L Y E D P R Q H S P G V L T D L R S A L V N N T I F A S L A V K Y D Y H K Y F K A V S P E L F H V I D D F V Q F Q L E K N E M Q G M D S E L R R S E E D E E K E E D I E V P K A M G D I F E S L A G A I Y M D S G M S L E T V W Q V Y Y P M M R P L I E K F S A N V P R S P V R E L L E M E P E T A K F S P A E R T Y D G K V R V T V E V V G K G K F K G V G R S Y R I A K S A A A R R A L R S L K A N Q P Q V P N S

  2. Open RaptorX: http://raptorx.uchicago.edu/
  3. Click on the “Submit” button in the first server option “RaptorX structure prediction”, or directly paste this uniform resource locator (URL) in the space for the URL: http://raptorx.uchicago.edu/StructurePrediction/predict/
  4. Type your job identification and email in the respective spaces, as shown in the previous chapter (Figure 30.1).
  5. Paste the input sequence (in FASTA format) in the sequence box under “Sequences for Prediction”.

31.4 JOB STATUS

Once the job has been started, the status can be seen in real time on the screen as below:

Job progress box (arrowed) of RaptorX after job submission.

FIGURE 31.1 Job progress box of RaptorX after job submission.

31.5 OUTPUT AND INTERPRETATION OF RESULTS

Open the specified email ID to get the results:

  1. Predicted structure in *.png format.
  2. Predicted structure in *.pdb format.
  3. Ligand binding site detail as *domain_pocket.txt file.

Click on the link provided in the mail to open the detailed result in a new window.

31.5.1 Section I: input sequence and domain partition

The whole protein sequence is partitioned into domains (here six domains), depending on the available template structures in PDB. The domains are indicated by assigning a domain number in every third row of sequence blocks.

Image described by caption.

FIGURE 31.2 Results windows of RaptorX, indicating assignment of protein domain and 3D prediction results.

31.5.2 Secondary structure prediction

The proportions of secondary structures (e.g., (Helix (H), Beta‐sheet (E) and Loop (L)) are given as percentages. In our study, the result shows “Secondary struct: 40%H, 10%E, 49%C”.

31.5.3 Solvent accessibility (ACC)

Three grades of solvent accessibility are designated by Buried (B: cut‐off value 10%), Medium (M: range 10–42%) and Exposed (E: cut‐off value: 42%). The present example shows: “Solvent access: 32%E, 38%M, 29%B”.

31.5.4 Quality assessment of predicted tertiary structure

Many parameters are checked to determine the quality of the predicted structure.

31.5.5 Probability of obtaining a random structure instead of the best one (P value)

The smaller the value, the better the prediction. If the P‐value is more than 10E‐3 and 10E‐4 for alpha and beta proteins respectively, the predicted sequence is to be discarded.

31.5.6 Alignment score (score)

The higher the score, the better is the prediction.

31.5.7 Number of identical residues (uSeqId and SeqId)

This indicates the number of identical residues in the alignment. Here, the “u” of “uSeqId” stands for un‐normalized. The higher the value, the better it is considered to be. In general, for a peptide sequence of 200 residues, the cut‐off value of normalized uSeqId is 30%.

31.5.8 Global distance test (GDT)

This estimates the modeling error from a score determined from the residues with modeling error.

31.5.9 Pocket multiplicity

This parameter is considered for binding site prediction in order to assess the quality of predicted pocket. Multiplicity higher than 40% is considered good.

31.6 QUESTIONS

  1. 1. Predict the tertiary structure of the caprine beta‐defensin (GenBank Acc. No. ABF71365.1) using an ab initio approach
  2. 2. Predict the tertiary structure of the same peptide (caprine beta‐defensin) by homology modeling, using a human homolog.
  3. 3. Can you compare the results obtained from the above two methods?
  4. 4. What parameters are considered in RaptorX for determining the goodness of the predicted structure?
..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset
3.138.34.52