Fabry:Sequence-based mutation analysis

From Bioinformatikpedia
Revision as of 11:04, 18 June 2012 by Rackersederj (talk | contribs) (Substitution matrices)

Fabry Disease » Sequence-based mutation analysis


The following analyses were performed on the basis of the α-Galactosidase A sequence. Please consult the journal for the commands used to generate the results.

Dataset preparation

Q279E
N215S
I289V
S65T
R356W
V316I
P323T
P40S
R118H
A143T


Amino acid properties

<figtable id="tab:aaProp"> Physicochemical properties of the chosen SNPs and changes of properties between wildtype (wt) and mutant (mt).
Used abbreviation in this table:
AA: Amino Acid, Pol: Side-chain polarity, Charge: Side-chain charge at pH 7.4, HI: Hydropathy index, RM: Residue Mass, iP: isoelectric point

SNP wt
AA
wt
Pol
wt
Charge
wt
HI
wt
RM
wt
iP
mt
AA
mt
Pol
mt
Charge
mt
HI
mt
RM
mt
iP
change in Pol change in Charge change in HI change in RM change in iP
Q279E Q polar neutral -3.5 128.131 5.65 E polar negative -3.5 129.116 3.15 none neutral to
negative
0 0.99 -2.5
N215S N polar neutral -3.5 114.104 5.41 S polar neutral -0.8 87.078 5.68 none none 2.7 -27.026 0.27
I289V I nonpolar neutral 4.5 113.160 6.05 V nonpolar neutral 4.2 99.133 6.00 none none -0.3 -14.027 -0.05
S65T S polar neutral -0.8 87.078 5.68 T polar neutral -0.7 101.105 5.60 none none 0.1 14.027 -0.08
R356W R polar positive -4.5 156.188 10.76 W nonpolar neutral -0.9 186.213 5.89 polar to
nonpolar
positive to
neutral
3.6 30.025 -4.87
V316I V nonpolar neutral 4.2 99.133 6.00 I nonpolar neutral 4.5 113.160 6.05 none none 0.3 14.027 0.05
P323T P nonpolar neutral -1.6 97.117 6.30 T polar neutral -0.7 101.105 5.60 nonpolar to
polar
none 0.9 3.988 -0.7
P40S P nonpolar neutral -1.6 97.117 6.30 S polar neutral -0.8 87.078 5.68 nonpolar to
polar
none 0.8 -10.039 -0.62
R118H R polar positive -4.5 156.188 10.76 H polar pos(10%),
neutr(90%)
-3.2 137.142 7.60 none positive to pos(10%),
neutr(90%)
1.3 -19.046 -3.16
A143T A nonpolar neutral 1.8 71.079 6.01 T polar neutral -0.7 101.105 5.60 nonpolar to
polar
none -2.5 30.026 -0.41

</figtable>



The polarity of the side chain determines whether an amino acid is hydrophobic or not. Hydrophobicity is a measure of how soluble an amino acid is in water. Hydrophobic amino acids are more likely to be found inside a protein, while hydrophilic amino acids rather are in contact with the aqueous environment. <ref>Hydrophobicity Index for Common Amino Acids http://www.sigmaaldrich.com/life-science/metabolomics/learning-center/amino-acid-reference-chart.html#hydro, June 16, 2012</ref> Therefore, depending on the localisation of an amino acid, a change in the polarity due to a mutation can cause a major defect. This may be the case in the SNPs P40S, A143T, P323T and R356W.
Furthermore the type of charge (positive or negative) is important for the structure of a protein, because it controls the binding of the amino acid to close-by residues. A modification again can break the coherence of the protein, which might happen when the mutations R118H, Q279E and R356W occur.
The hydropathy index of an amino acid is a number representing the hydrophobic or hydrophilic properties of its sidechain.<ref>Kyte J, Doolittle RF (May 1982). A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157 (1): 105–32. PMID 7108955. </ref> The larger the number is, the more hydrophobic the amino acid, thus the most hydrophobic amino acid is isoleucine (4.5) and the most hydrophilic one is arginine (-4.5). Considering a hydropathy change greater than 1 (in both direction, positive and negative) as crucial, only one SNP highly increases the hydrophobicity (A143T) and 3 increase the hydrophilic character of the position (R118H, N215S and R356W)
The average residue mass ranges from 57.052 (Glycine) to 186.213 (Tryptophan), thus we expect an alteration of the mass of greater than 10 as critical. This concerns all mutations expect for Q279E and P323T.
The isoelectric point is the pH at which an amino acid carries no net charge. Below the pI it carries a net positive charge, above it a net negative charge. Since the pH in the human body is on average 6.7, only three amino acids are positively charged (Histidine, Lysine and Arginine). The pI ranges from 2.85 (Aspartic acid) to 10.76 (Arginine), therefore we considered a change of 0.8 as probably desease causing. This applies only for R118H, Q279E and R356W.

Simple structural analysis

  • Now take into consideration where in the protein the mutation occurs and document: Create a picture with PyMOL showing the original and mutated residue in the protein. Use PyMOL for this. More thorough structural analyses will be introduced in the next task.

Secondary Structure

<figtable id="tab:Location"> Secondary structure assignment predicted by the three methods Psipred, Reprof and DSSP in Task 3 for the mutated amino acid itself and
the ten adjacent residues to the left and to the right.
H represents a helix at this position, C represents coiled regions, E sheets and - is a not predictable region

SNP SecStruc
Psipred
SecStruc Psipred
long
SecStruc
Reprof
SecStruc Reprof
long
SecStruc
DSSP
SecStruc DSSP
long
Q279E H CCCCCCCHHHHHHHHHHHHHH H EECCCCCCHHHHHHHHHHHHH H CCCCC--HHHHHHHHHHHHHC
N215S H CCCCCCCCCCHHHHHCCCCCC H CECCCCCCCCHHHHHHHHHHH H HHHCCCC---HHHHCCC-CEE
I289V H HHHHHHHHHHHCCCEEEECCC H HHHHHHHHHHHHHCHHCCCCC C HHHHHHHHHHCC--EEE-C-C
S65T H CCCCCCCCCCHHHHHHHHHHH H CCCCCCCHHHHHHHHHHHHHH H CCC-CCCC-CHHHHHHHHHHH
R356W C CCEEEEEEECCCCCCEEEEEE C HHHHHHHHHHCCCCCCCCHHH - CEEEEEEEE---CCC-EEEEE
V316I H HHHCCCCHHHHHHCCCCCCCC E HHHHCCCCCEEEECCCCCCCC H HHHHHH-HHHHHHHC-CC---
P323T C HHHHHHCCCCCCCCCEEEEEC C CCEEEECCCCCCCCCCEECCC C HHHHHHHC-CC----EEEE-C
P40S C CCCCCCCCCCCCCCCCCCCCC C HHCCCCCCCCCCCHHHHHHEE - ---CC--CC--EEEECHHHHC
R118H H CCCCCCCCHHHHHHHHHHCCC H CCCCCCHHHHHHHHHHHCCCC H -CCC-CCHHHHHHHHHHHCC-
A143T C EECCCCCCCCCCCCCCCCHHH C EEECCCCCCCCCCCCCCCCCC C EEECCCE-CCCCE--CCCHHH

</figtable>

Despite the fact, we know from last weeks' task, that the disease causing mutations are spread all over the protein without any respect to the secondary structure, we assumed we had no prior knowledge about it. Thus we looked at the predicted secondary structure at the position of each point mutation and its surrounding (10 residues to the left and 10 residues to the right). The only remarkable fact is, that there are (almost) no sheets at the mutated residues. From Task 6 we know, that this happened only by chance and due to the small amount of picked SNPs.

Substitution matrices

<figtable id="tab:Subsmatr"> Substitution values for all SNPs,
assigned by the three substitution matrices
BLOSUM62, PAM1 and PAM250.

SNP Value
BLOSUM62
Value
PAM1
Value
PAM250
Q279E 3 27 2
N215S 1 20 1
I289V 4 33 4
S65T 2 38 1
R356W -4 8 2
V316I 4 57 4
P323T -2 4 0
P40S -1 12 1
R118H 0 10 2
A143T 0 32 1

</figtable>

Since the PAM1 and the PAM250 matrices are designed for proteins of very diverse degree of kinship, 99% and ~20% relationship, respectively, those two matrices tend to give contradictory scores of how likely a substitution is.

PSSM

  • Getting a bit closer to evolution you will have to create a PSSM (position specific scoring matrix) for your protein sequence using PSI-BLAST (5 iterations). How conserved are the WT residues in your mutant positions? How is the frequency of occurrence (conservation) for the mutant residue type? Anything interesting?

Multiple sequence alignment

  • And another step close to evolution: Identify all mammalian homologous sequences. Create a multiple sequence alignment for them with a method of your choice. Using this you can now calculate conservation for WT and mutant residues again. Compare this to the matrix- and PSSM-derived results.

Scoring methods

SIFT

<figtable id="tab:Sift"> Sift Scores

SNP Prediction Sift Score Sequences represented at this position
P40S AFFECT PROTEIN FUNCTION 0.00 41
S65T AFFECT PROTEIN FUNCTION 0.01 45
R118H be TOLERATED 0.06 48
A143T AFFECT PROTEIN FUNCTION 0.01 48
N215S AFFECT PROTEIN FUNCTION 0.01 48
Q279E AFFECT PROTEIN FUNCTION 0.00 48
I289V AFFECT PROTEIN FUNCTION 0.05 48
V316I be TOLERATED 0.75 48
P323T AFFECT PROTEIN FUNCTION 0.01 48
R356W AFFECT PROTEIN FUNCTION 0.01 47

</figtable>

Median sequence conservation: 2.99

SIFT

Polyphen2

<figtable id="tab:Polyphen"> Polyphen Scores

SNP rs ID Sec Struc Prediction pph2 Class pph2 Prob pph2 FPR pph2 TPR pph2 FDR
Q279E rs28935485 H probably damaging deleterious 0.983 0.0387 0.745 0.0657
N215S rs28935197 . benign neutral 0.048 0.167 0.941 0.194
I289V ? H probably damaging deleterious 0.975 0.0436 0.762 0.072
S65T ? . probably damaging deleterious 0.995 0.0277 0.681 0.0521
R356W ? . probably damaging deleterious 1 0.00026 0.00018 0.0109
V316I ? H benign neutral 0.308 0.113 0.904 0.144
P323T ? T possibly damaging deleterious 0.612 0.091 0.872 0.124
P40S ? . probably damaging deleterious 1 0.00026 0.00018 0.0109
R118H ? H benign neutral 0.015 0.209 0.956 0.229
A143T ? T probably damaging deleterious 1 0.00026 0.00018 0.0109

</figtable>

Polyphen2 [1]

SNAP

    • SNAP is installed on the VirtualBox and should be used command-line only. -- As blast is the bottleneck of SNAP, and you are doing that anyway, we might as well look at all possible substitutions in the position of our mutations. This way we can learn much more about the nature of the given mutation: Is our mutation problematic because we introduce an unwanted effect, or because the WT residue is essential and by mutating we remove that?


Results and Conclusion

<figtable id="tab:Overview"> This table gives an overview over all features examined in the sections above. The red background color indicates a disease causing prediction, the green color a non-disease causing one. In the end all red fields are summed up for each row and the resulting value leads to our prediction given in <xr id="tab:Result"/>.
Used abbreviation in this table:
AA: Amino Acid, Pol: Side-chain polarity, Charge: Side-chain charge at pH 7.4, HI: Hydropathy index, RM: Residue Mass, iP: isoelectric point

SNP change in Pol change in Charge change in HI change in RM change in iP SecStruc Psipred SecStruc Reprof SecStruc DSSP Value
BLOSUM62
Value
PAM1
Value
PAM250
Sift Score pph2 Class Sum
bad scores
A143T nonpolar to
polar
none -2.5 30.026 -0.41 C C C 0 32 1 0.01 deleterious 6
R356W polar to
nonpolar
positive to
neutral
3.6 30.025 -4.87 C C - -4 8 2 0.01 deleterious 9
I289V none none -0.3 -14.027 -0.05 H H C 4 33 4 0.05 deleterious 5
V316I none none 0.3 14.027 0.05 H E H 4 57 4 0.75 neutral 4
R118H none positive to pos(10%),
neutr(90%)
1.3 -19.046 -3.16 H H H 0 10 2 0.06 neutral 8
N215S none none 2.7 -27.026 0.27 H H H 1 20 1 0.01 neutral 7
Q279E none neutral to
negative
0 0.99 -2.5 H H H 3 27 2 0.00 deleterious 7
P40S nonpolar to
polar
none 0.8 -10.039 -0.62 C C - -1 12 1 0.00 deleterious 7
S65T none none 0.1 14.027 -0.08 H H H 2 38 1 0.01 deleterious 7
P323T nonpolar to
polar
none 0.9 3.988 -0.7 C C C -2 4 0 0.01 deleterious 6

</figtable>

<figtable id="tab:Result"> All examined SNPs with the "sum bad score" according to <xr id="tab:Overview"/> and our resulting
prediction. The table also shows if the SNP truly is disease causing or not and whether our sequence
based prediction is true

SNP Sum
bad scores
Prediction True classification Result prediction
A143T 6 Non-disease causing Disease causing Wrong
R356W 9 Disease causing Disease causing Right
I289V 5 Non-disease causing Non-disease causing Right
V316I 4 Non-disease causing Non-disease causing Right
R118H 8 Disease causing Non-disease causing Wrong
N215S 7 Disease causing Disease causing Right
Q279E 7 Disease causing Disease causing Right
P40S 7 Disease causing Disease causing Right
S65T 7 Disease causing Disease causing Right
P323T 6 Non-disease causing Non-disease causing Right

</figtable>

In <xr id="tab:Overview"/> we list all afore gathered information in condensed form and highlight values that we consider as an indicator for a disease causing mutation with red color. Results contained in green colored fields are considered neutral. The summed up score of disease indicators is again shown in <xr id="tab:Result"/> along with our prediction for each SNP. Mutations with score smaller than 7 are considered neutral, greater or equal to that disease causing. Next to the prediction we show the true classification of the single nucleotide polymorphism acording to the mapping we did in Task 6. We show that only two of the ten predictions are wrong, which we consider as a surprisingly good result.



References

<references/>