Supplementary Materials
RNAsnp: Efficient detection of local RNA secondary structure
changes induced by SNPs
Radhakrishnan Sabarinathan, Hakim Tafer, Stefan E. Seemann,
Ivo L. Hofacker, Peter F. Stadler and Jan Gorodkin
1Sabarinathan et al., Human Mutation
Figure S1 - The graphical representation illustrates the base pairing probability matrix of a sequence. Each
cell [i,j] contains the base pairing probability of i with j, i.e. Pij . We employed a recursive method to add all
the probabilities along the row, which helps for the faster computation of position-wise pairing probabilities
(pi). Consider [k, l] is a sequence interval k < i < l, and denotes Mi,k and Ni,l the probabilities that i has
a pairing partner in the interval [k, i − 1] and the interval [i + 1, l], resp. These auxiliary variables satisfy
Mi,k = Mi,k−1 + Pki and Ni,l = Ni,l+1 + Pil. Obviously, for all k < i < l we have pii[k, l] = Mi,k + Ni,l.
Also, we can directly compute the expected number of base pairs inside and outside the substructure, Jkl
and Ekl, resp.
2Sabarinathan et al., Human Mutation
0.0
0.2
0.4
0.6
0.8
1.0
X
X X
X
X
1 2 3 4 5
alpha
p−v
alue
X Mean
Figure S2 - The data set of 30 SNPs with reported structural effects was used to test the effect of different
α values ranging from 1 to 5 in steps of 1. The α parameter determines the ratio between the expected
number of base pairs inside (Jkl) a local interval [k, l], compared to the expected numbers of base pairs (Ek,l)
cross the boundaries of the interval. The box plot shows the distribution of p-values for the 30 known SNPs
changes for the different α values. As expected, the increase in mean p-value correlate with increase in the α
value, because the higher the alpha value the greater the increase in expected number of base pairs inside the
[k, l] compare to the outside. This eventually results in the selection of larger interval [k, l] for comparison
and thus the measure of local structural effect become less significance. Thus, the α=1 is chosen as default.
3Sabarinathan et al., Human Mutation
Figure S3 - Schematic representation of RNAplfold base pairing probability matrix calculated for a window
of 400 nts centering the SNP position. The region highlighted in dark gray is used for the initial screen to
find the position k that maximizes d2(k)(P, P ∗). The local window (LW) ranges from k to k + h′ + h′′, where
h′ = 20 and h′′ = 120. The LW contains the interval [u, v] from which the optimized distance measure (d♯)
is obtained.
d(P, P∗)
0.0 0.4 0.8 0.0 0.4 0.8 0.0 0.4 0.8
0.0
0.4
0.8
0.0
0.4
0.8
0.97 r(P, P∗)
0.98 0.97 d(pi, pi∗)
0.0
0.4
0.8
0.0
0.4
0.8
0.98 0.99 0.98 r(pi, pi∗)
0.99 0.98 0.98 0.97 d(ξ< >, ξ < > ∗)
0.0
0.4
0.8
0.0
0.4
0.8
0.97 0.98 0.99 0.97 0.98 r(ξ< >, ξ< > ∗)
0.0 0.4 0.8
0.68 0.65
0.0 0.4 0.8
0.65 0.67
0.0 0.4 0.8
0.66 0.62
0.0 0.4 0.8
0.0
0.4
0.8
δ
Figure S4 - The correlation between global(dis)similarity measures was computed using the data set of
7000 random sequences of length 400 and considered SNPs at 200 position. The measures distance(d) and
correlation coefficient(r) computed on various base pairing probabilities (P - full base pairing probabilities,
pi - position-wise pairing probabilities, ξ<> - position-wise, distinguished up- and down-stream paring prob-
abilities), correlate with each other. However, these measures does not correlate well with the Euclidean
distance (δ) computed between the distribution of wild-type and mutant structures.
4Sabarinathan et al., Human Mutation
0.0 0.2 0.4 0.6 0.8 1.0
0.0
0.2
0.4
0.6
0.8
1.0
0.00 0.04 0.08
0.00
0.04
0.08
0.0 0.2 0.4 0.6 0.8 1.0
0.0
0.2
0.4
0.6
0.8
1.0
0.00 0.04 0.08
0.00
0.04
0.08
0.0 0.2 0.4 0.6 0.8 1.0
0.0
0.2
0.4
0.6
0.8
1.0
0.00 0.04 0.08
0.00
0.04
0.08
0.0 0.2 0.4 0.6 0.8 1.0
0.0
0.2
0.4
0.6
0.8
1.0
0.00 0.04 0.08
0.00
0.04
0.08
Figure S5 - Comparison of rank-based p-values and fitted p-values for a set of 5000 random numbers. The
p-values are calculated from the background distribution of length 400nts, G+C content between 50 and
60% and the SNP position at 200 position. In all four cases, the comparison of rank-based p-value versus
the fitted p-value shows high correlation (r>0.9). The inset figure shows the comparison of p-values which
are less than 0.1.
0.0 0.2 0.4 0.6
0
2
4
6
8
10
12
Effect of different minimum lengths
Distance
Density
minLen
20
30
40
50
60
70
80
90
100
−1.0 −0.5 0.0 0.5 1.0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Effect of different minimum lengths
Correlation Coefficient
Density
minLen
20
30
40
50
60
70
80
90
100
Figure S6 - The effect of minimum length on dmax/RNAfold and rmin/RNAfold was tested with different
cut-off values. In both cases, the length cut-off 50 shows the unimodal distribution.
5Sabarinathan et al., Human Mutation
0.0
0.2
0.4
0.6
0.8
1.0
0.0
0.2
0.4
0.6
0.8
1.0
d
#
R
NAp
lfold
r #
R
NAp
lfold
c.[51A>G;..]
n.[8953C>A;..]
c.−12C>G c.−867G>T c.−697G>C
c.36C>T
c.−764G>A c.903T>C c.*310T>A c.[*26A>T
;..]
c.−853C>G c.*297A>C c.*260C>G c.*102A>C n.190T>G c.1007G>C n.255C>G c.*1330C>G n.8546G>A
c.−52A>G c.−149G>C c.*218C>T
n.15G>A n.229G>A c.−798G>C
c.*72A>G c.455T>C c.−29G>C c.*463T>G c.*1969T>C
SNPs
p−v
alue
Figure S7 - The significance of structural effects as predicted by d♯/RNAplfold and r♯/RNAplfold for the
30 known SNPs. The p-values are shown as bars and the red dashed line represents the selected threshold
value 0.1. The four experimentally validated examples are indicated in green. The SNPs were described
according to HGVS nomenclature.
6Sabarinathan et al., Human Mutation
MFE
bp_prob
0.0
0.2
0.4
0.6
0.8
1.0
d_max/RNAfold
MFE
bp_prob
0.0
0.2
0.4
0.6
0.8
1.0
r_min/RNAfold
p = 0.29 p = 0.33
Figure S8 - The data set of 30 SNPs with reported structural effects was used to compare the local measures
dmax and rmin based on base pairing probability of structural ensemble and the base pairs of minimum free
energy (MFE) structure, which was calculated using RNAfold. The box plot shows the distribution of
p-values for the 30 known SNPs obtained for the two measures. The p-values from the local measures
based on the structural ensemble are in general smaller than the p-values derived from the MFE structure.
The comparison of two p-value distributions using Wilcoxon rank sum test, however, shows no significant
difference (P>0.2) for either of the dmax or rmin measures. This may be explained by the fact that the
available data set SNPs are small.
7Sabarinathan et al., Human Mutation
M
AF>40% MAF<1%
disease−SNPs
0.0
0.2
0.4
0.6
0.8
1.0
d_max/RNAfold
MAF>40% MAF<1%
disease−SNPs
0.0
0.2
0.4
0.6
0.8
1.0
r_min/RNAfold
MAF>40% MAF<1%
disease−SNPs
0.0
0.2
0.4
0.6
0.8
1.0
SNPfold
p = 0.279
p = 0.937
p = 0.741 p = 0.335
p = 0.556 p = 0.678
Figure S9 - The box plots show the distribution of p-values calculated for three different data sets that are
equal in number of SNPs (n=501) obtained from a) HapMap database with minor allele frequency greater
than 40%, b) dbSNP (build 135) with minor allele frequency less than 1% and c) disease-associated SNPs
from Human Gene mutation Database (HDMD). The structural effect of these SNPs were predicted with
RNAsnp (Mode 1) and SNPfold [Halvorsen et al., 2010]. For each structural (dis)similarity measure, the
difference between the p-value distribution of the three data sets were compared using Wilcoxon rank sum
test. It shows that P-value of the wilcoxon rank sum test are higher than the significant level of 0.01 and
suggests that there is no significant difference between the p-value distributions of three different SNP data
sets. The same scenario was observed in the case of global measure, SNPfold.
8Sabarinathan et al., Human Mutation
Table S1 - An overview of 30 SNPs with reported structural effect on RNA secondary structure. The SNPs
were described according to HGVS nomenclature.
Disease/Phenotype Gene Refseq SNP Validation Reference
Alteration of RNA replication
in HCV
NS5B AJ238799.1 n.[8953C>A;8955T>G] experimental [1]
Tumor formation p53 NM 001126114.2 c.[51A>G;54A>C;57T>C] experimental [2]
HIV-1 resistance against RNAi Nef K02013.1 n.8546G>A experimental [3]
Alteration of alanyl tRNA
synthetase expression in human
AARS D32050.1 c.903T>C experimental [4]
Cowden Syndrome PTEN NM 000314.4 c.-867G>T predicted [5]
c.-853C>G predicted
c.-798G>C predicted
c.-764G>A predicted
Occult-hepatitis B virus
infection
- EU155893.1 n.190T>G predicted [6]
Psychiatric disorders TPH2 NM 173353.3 c.-52A>G predicted [7]
Diurnal preference PER2 NM 022817.2 c.-12C>G predicted [8]
Nasopharyngeal
Carcinoma Risk
TLR4 NM 138554.4 c.1007G>C predicted [9]
Alteration in localization
of rat MT1 mRNA
MT1 NM 138826.4
c.[*26A>T;*27G>C;
28G>C;*29T>A;*30G>C]
predicted [10]
Cone Dystrophy PDE6H NM 006205.2 c.-29G>C predicted [11]
Congenital
heart disease
GATA4 NM 002052.3 c.*260C>G predicted [12]
c.*218C>T predicted
Antipsychotic induced
weight gain
HTR2C NM 000868.2 c.-697G>C predicted [13]
Muscular dystrophies SGCG U34976.1 c.*102A>C predicted [14]
Pain sensitivity COMT NM 007310.2 c.36C>T predicted [15]
Alteration of plasma
zymogen TAFI
concentration
CPB2 NM 001872.3 c.*310T>A predicted [16]
c.*72A>G predicted
Anauxetic dysplasia RMRP NR 003051.3 n.255C>G predicted [17]
n.15G>A predicted
Mental retardation CDK5R1 NM 003885.2 c.*1330C>G predicted [18]
Alteration of RNA
translation in HCV-1b
IRES EU857431.1 n.229G>A predicted [19]
Hyperferritinaemia
cataract syndrome
FTL NM 000146.3 c.-149G>C predicted [20]
Resistance to
Hirschsprung disease RET NM 020975.4 c.*1969T>C predicted [21]
Ischemic
Cardiomyopathy
ADORA1 NM 000674.2 c.*297A>C predicted [22]
Neuropsychiatric
disorders SLC6A4 NM 001045.4 c.*463T>G predicted [23]
Cutaneous melanoma BMP4 NM 001202.3 c.455T>C predicted [24]
9Sabarinathan et al., Human Mutation
Results of Rchange analysis
In contrast to the RNA mutation analysis programs (like RNAsnp) based on base pairing probability, the
program Rchange [Kiryu and Asai , 2012] was recently developed to predict the structural effect based on
the changes in the energy of RNA secondary structures in response to the single or double mutations. This
program was tested on our data set of four known SNPs whose structural effect has been experimentally
verified. Since, the program can handle either single or double mutants, only three out of the four known
SNPs were tested successfully. Rchange computed the changes in thermodynamic entropy (S), mean energy
(U) and ensemble free energy (F ) between the wild-type and mutant RNA secondary structures. In order
to compute the significance value, the RNA sequence of each SNP was shuffled ten times and subjected
each of them with Rchange to compute the energy difference for the random mutations. Using this result as
background distribution, the p-value is estimated using a non-parametric approach for the results of known
SNPs (Table S2). Table S2 shows that the RNAsnp predicted only one SNP (n.8546G>A) have high structural
effect based on the dS/S and dF/|F | measures.
Table S2 - The results of Rchange for the three out of four SNPs with reported structural effect. The
measures dS/S, dU/|U | and dF/|F | represents, respectively, the difference in the thermodynamic entropy of
RNA secondary structures, mean energy and ensemble free energy between the wild-type and mutant RNAs.
The p-value with less than 0.1 significance level are highlighted with bold text.
Rchange RNAsnp
dS/S dU/|U| dF/|F | dmax/RNAfold rmin/RNAfold
Ref Gene Accession SNP (p-value) (p-value) (p-value) (p-value) (p-value)
1 NS5B AJ238799.1 n.[8953C>A;8955T>G] 0.911 0.802 0.729 0.074 0.121
2 AARS D32050.1 c.903T>C 0.327 0.317 0.303 0.072 0.093
3 Nef K02013.1 n.8546G>A 0.050 0.590 0.089 0.094 0.083
10Sabarinathan et al., Human Mutation
Table S3 - List of disease associated SNPs from HGMD that are predicted to have significant local structural
effect (p-value < 0.1) by dmax/RNAfold or rmin/RNAfold of RNAsnp (Mode 1). The SNPs were described
according to HGVS nomenclature.
HGMD Genbank p-value
Disease/phenotype Gene Accession Accession SNP dmax/RNAfold rmin/RNAfold
Pseudohypoaldosteronism NR3C2 CR030126 NM 000901.4 c.-2C>G 0.017 0.022
Hypertension EDN2 CR994679 NM 001956.3 c.*390G>A 0.036 0.021
Obesity CNR1 CR073542 NM 033181.3 c.*2394A>G 0.032 0.036
Myocardial infarction GP1BA CR022116 NM 000173.5 c.-5T>C 0.040 0.037
Colorectal cancer INSR CR082021 NM 001079817.1 c.*104A>G 0.042 0.030
Graves’ disease FCRL3 CR067134 NM 052939.3 c.-11G>C 0.011 0.042
Increased triglyceride levels ABCA1 CR025352 NM 005502.3 c.-279C>G 0.044 0.022
Insulin resistance hypertension RETN CR032443 NM 020415.3 c.*62G>A 0.045 0.043
Cartilage-Hair hypoplasia RMRP CR063417 NR 003051.3 n.215A>G 0.048 0.027
Hypercholesterolaemia LDLR CR971948 NM 000527.4 c.-14C>A 0.025 0.048
Glaucoma CYP1B1 CR032431 NM 000104.3 c.-286C>T 0.063 0.036
Reduced transcriptional activity NR3C1 CR016150 NM 001024094.1 c.-219C>A 0.044 0.063
HDL cholesterol levels LIPG CR032437 NM 006033.2 c.*482A>G 0.051 0.065
Factor VII deficiency F7 CR090334 NM 019616.2 c.-44T>C 0.066 0.042
HaemophiliaA F8 CR070421 NM 000132.3 c.-112G>A 0.074 0.010
Cartilage-Hair hypoplasia RMRP CR064472 NR 003051.3 n.10T>C 0.076 0.024
VonHippel-Lindau syndrome VHL CR011856 NM 000551.3 c.*7C>G 0.076 0.065
Obesity SLC6A14 CR035766 NM 007231.3 c.*178C>G 0.078 0.062
Spasticparaplegia31 REEP1 CR082030 NM 022912.2 c.*14C>T 0.033 0.081
Hyperferritinaemia-cataract
syndrome
FTL CR061334 NM 000146.3 c.-178T>G 0.052 0.097
Severe iron overload ALAS2 CR090059 NM 001037968.3 c.-69C>T 0.078 0.390
Systemic lupus erythematosus CRP CR040151 NM 000567.2 c.*1082G>A 0.048 0.316
Migraine EDNRA CR011854 NM 001957.3 c.-67G>A 0.053 0.106
Hyperferritinaemia-cataract
syndrome FTL HR030029 NM 000146.3 c.-171C>G 0.100 0.354
Cholesterol level GHRL CR065638 NR 024132.1 n.316G>C 0.056 0.110
Colorectal cancer MLH1 CR033148 NM 000249.3 c.-28A>T 0.097 0.140
Panencephalitis MX1 CR040301 NM 002462.3 c.-434G>T 0.058 0.258
Lipoprotein/Triglyceride levels PCK1 CR054265 NM 002591.3 c.*431T>C 0.084 0.103
Cowden disease PTEN CR032094 NM 000314.4 c.-930G>A 0.019 0.109
Diabetes PTEN CR033149 NM 000314.4 c.-8C>G 0.098 0.266
Chronic obstructive pulmonary
disease
SERPINA1 CR061339 NM 001127701.1 c.-458C>T 0.079 0.129
Haemochromatosis SLC40A1 CR057017 NM 014585.5 c.-187A>G 0.045 0.282
11Sabarinathan et al., Human Mutation
Table S3 - continued
HGMD Genbank p-value
Disease/phenotype Gene Accession Accession SNP dmax/RNAfold rmin/RNAfold
Aplastic anaemia TERC CR057475 NR 001566.1 n.117A>C 0.035 0.223
Aplastic anaemia TERC CR080776 NR 001566.1 n.2G>C 0.082 0.198
Chondrocalcinosis ANKH CR057902 NM 054027.4 c.-4G>A 0.131 0.029
Factor XI deficiency F11 CR064469 NM 000128.3 c.-54G>A 0.124 0.069
Factor VII deficiency F7 CR002894 NM 019616.2 c.-30A>C 0.150 0.076
IPEX syndrome FOXP3 CR063404 NM 001114377.1 c.-7G>T 0.116 0.067
Decreased expression GCH1 CR075245 NM 000161.2 c.*243C>T 0.148 0.065
Frontotemporal dementia? GRN CR072310 NM 002087.2 c.-72G>T 0.178 0.054
Hypercholesterolaemia LDLR CR042574 NM 000527.4 c.-153C>T 0.116 0.059
Hypercholesterolaemia LDLR CR951555 NM 000527.4 c.-138T>C 0.170 0.091
Cellular response to cadmium MT2A CR066330 NM 005953.3 c.-77A>G 0.126 0.053
Reduced expression NEIL2 CR085800 NM 145043.2 c.-586C>G 0.126 0.091
Schizophrenia NOS1 CR025919 NM 000620.4 c.*276C>T 0.181 0.059
Decr.serum leptin levels
in lean indiv.
POMC CR035490 NM 000939.2 c.*63C>T 0.227 0.092
Hirschsprung disease RET CR951557 NM 020975.4 c.-27C>G 0.173 0.095
Cartilage-Hair hypoplasia RMRP CR012677 NR 003051.3 n.263G>T 0.130 0.051
RMRP CR021393 NR 003051.3 n.183G>C 0.135 0.085
RMRP CR021394 NR 003051.3 n.212C>G 0.182 0.067
RMRP CR054268 NR 003051.3 n.183G>T 0.136 0.085
RMRP CR054277 NR 003051.3 n.214C>G 0.108 0.039
Pancreatitis SPINK1 CR001469 NM 003122.3 c.-53C>T 0.121 0.020
Nasopharyngeal cancer TLR4 CR068105 NR 024168.1 n.3938G>C 0.210 0.083
12Sabarinathan et al., Human Mutation
Table S4 - List of disease associated SNPs from GWASdb that are predicted to have significant local
structural effect by dmax/RNAplfold (p < 0.4) and dmax/RNAfold (p < 0.1) of RNAsnp (with mode 3). The
SNPs were described according to HGVS nomenclature.
Ensembl p-value
Disease/phenotype id UTR dbSNP dmax/RNAfold rmin/RNAfold
Suicide attempts in
bipolar disorder
ENST00000373055 3 rs7822:T>C 0.0290 0.0199
Lapatinib-induced
hepatotoxicity ENST00000360403 5 rs489676:C>G 0.0391 0.0925
Alzheimer‘s disease (late onset) ENST00000368485 3 rs7514452:C>T 0.1847 0.0984
Multiple complex diseases ENST00000368476 3 rs11264221:C>T 0.1611 0.0356
Ischemic stroke;Stroke ENST00000329117 3 rs11360:A>G 0.0594 0.0514
Systemic lupus erythematosus ENST00000255030 3 rs1205:G>A 0.2241 0.0477
Suicide attempts in
bipolar disorder
ENST00000333360 3 rs5357:T>C 0.0896 0.0296
Alcohol dependence ENST00000319387 3 rs4233175:A>G 0.3642 0.0883
Sudden cardiac arrest ENST00000260585 3 rs3820937:G>C 0.2935 0.0826
GWAS of height-adjusted
highest forced expiratory volume
in a British population
ENST00000379066 3 rs1056021:T>C 0.0021 0.0074
Urinary metabolites ENST00000426016 3 rs6704656:T>A 0.2238 0.0903
Amyotrophic Lateral
Sclerosis (ALS) ENST00000306448 5 rs896210:C>T 0.0394 0.0739
Multiple complex diseases ENST00000306503 3 rs17823065:T>C 0.0364 0.0249
Parkinson‘s disease;
Multiple complex diseases ENST00000254630 3 rs11395:T>C 0.3539 0.0370
Urinary metabolites ENST00000259213 3 rs4849142:C>T 0.0335 0.0690
Parkinson‘s disease ENST00000338983 3 rs8446:T>C 0.0482 0.0939
Multiple complex diseases ENST00000443029 5 rs2290536:T>G 0.3384 0.0222
Urinary metabolites ENST00000357632 3 rs17765088:C>G 0.0043 0.0108
Lung adenocarcinoma ENST00000433104 3 rs3172494:C>A 0.0015 0.0166
Serum calcium ENST00000344337 3 rs17201246:G>T 0.3573 0.0712
Multiple continuous traits
in DGI samples ENST00000305097 3 rs16864613:C>G 0.1268 0.0405
Alzheimer‘s disease (late onset) ENST00000337774 3 rs3821801:A>G 0.0123 0.0753
Serum uric acid;Serum urate ENST00000326756 3 rs3217:G>A 0.0439 0.0363
GWAS of systolic blood
pressure in a British population ENST00000344157 3 rs2293595:A>G 0.1178 0.0504
Serum uric acid ENST00000237596 3 rs2728121:C>T 0.0631 0.0182
Parkinson‘s disease ENST00000394989 3 rs3857053:G>A 0.1326 0.0566
Alcohol dependence ENST00000515683 3 rs2298753:A>G 0.0012 0.0072
Multiple complex diseases ENST00000285311 3 rs17509643:C>G 0.1033 0.0770
potassium response to spironolactone ENST00000355292 5 rs2070951:C>G 0.0466 0.0167
Alzheimer‘s disease (late onset) ENST00000061240 3 rs2279723:C>A 0.0611 0.0564
Alzheimer‘s disease (late onset) ENST00000284274 3 rs25952:A>C 0.1341 0.0659
Thrombosis ENST00000356834 3 rs1298:C>T 0.0129 0.0907
13Sabarinathan et al., Human Mutation
Table S4 - continued
Ensembl p-value
Disease/phenotype id UTR dbSNP dmax/RNAfold rmin/RNAfold
Common traits (Other) ENST00000380956 3 rs1050975:G>A 0.1541 0.0154
Phospholipid levels (plasma) ENST00000354666 3 rs4532436:G>C 0.0171 0.0486
Multiple complex diseases ENST00000383555 3 rs2073149:A>T 0.0943 0.0114
Lung adenocarcinoma;Rheumatoid Arthritis ENST00000376883 5 rs2535238:G>T 0.0419 0.0468
Rheumatoid arthritis;Lung cancer ENST00000449742 3 rs2257914:G>T 0.0303 0.0103
Rheumatoid Arthritis ENST00000375015 3 rs482194:T>C 0.0844 0.0795
Rheumatoid Arthritis ENST00000395388 5 rs14004:C>A 0.2745 0.0265
Serum metabolites;
Multiple complex diseases;
Multiple sclerosis
ENST00000395388 3 rs7194:G>A 0.3531 0.0417
HIV-1 control ENST00000374897 3 rs241454:T>C 0.1504 0.0461
Multiple complex diseases ENST00000374680 3 rs2744537:T>G 0.1538 0.0268
Multiple complex diseases ENST00000482399 3 rs461338:A>G 0.1284 0.0420
Prostatic Neoplasms;height ENST00000311565 5 rs2016520:C>T 0.2705 0.0659
Lipoprotein-associated
phospholipase A2 activity and mass ENST00000544460 3 rs12528857:C>A 0.3672 0.0645
Attention Deficit Disorder
with Hyperactivity ENST00000369838 3 rs1062793:A>G 0.0807 0.0925
Coronary heart disease; ENST00000367882 3 rs12190287:C>G 0.0345 0.0622
Bone mineral density (spine) ENST00000367290 3 rs6932603:T>C 0.1815 0.0227
Multiple complex diseases ENST00000222792 3 rs1420145:C>G 0.0090 0.0138
Aortic root size;Aortic root size ENST00000360415 3 rs875971:T>C 0.2875 0.0137
Major depressive disorder ENST00000005178 3 rs11531570:C>T 0.3857 0.0633
Alzheimer‘s disease ENST00000205402 3 rs4564:G>A 0.1477 0.0272
Glaucoma (primary open-angle) ENST00000222693 3 rs1052990:T>G 0.0436 0.0439
Inflammatory Bowel Diseases ENST00000466675 3 rs4721:A>C 0.0507 0.0936
Information processing speed ENST00000401878 3 rs2007922:G>A 0.0018 0.0169
Multiple complex diseases ENST00000256255 3 rs3189926:T>G 0.1546 0.0134
Multiple continuous traits
in DGI samples ENST00000342228 3 rs6841:C>G 0.0352 0.0934
Suicide attempts in bipolar disorder ENST00000289957 5 rs4950:G>A 0.0926 0.0633
Suicide attempts in bipolar disorder ENST00000521271 3 rs1044730:C>T 0.0089 0.0812
Amyotrophic Lateral Sclerosis (ALS) ENST00000297848 3 rs2429:G>T 0.1519 0.0903
Parkinson‘s disease ENST00000314393 3 rs3802266:A>G 0.3048 0.0906
Other erythrocyte phenotypes;
Multiple complex diseases ENST00000381750 3 rs1053872:G>C 0.0308 0.0516
Parkinson‘s disease (age of onset) ENST00000374193 3 rs10817478:A>G 0.0629 0.0215
Proinsulin levels ENST00000372155 3 rs306549:G>C 0.0282 0.0752
Achilles tendinopathy ENST00000371817 3 rs12722:C>T 0.1740 0.0536
14Sabarinathan et al., Human Mutation
Table S4 - continued
Ensembl p-value
Disease/phenotype id UTR dbSNP dmax/RNAfold rmin/RNAfold
Multiple complex diseases ENST00000446108 3 rs8463:A>G 0.0289 0.0987
Cognitive test performance;
Hirschsprung‘s disease ENST00000355710 3 rs17028:C>T 0.0768 0.0330
Urinary metabolites ENST00000333254 5 rs41386650:A>G 0.1587 0.0961
Type II Diabetes Mellitus ENST00000396952 3 rs10500609:A>G 0.0848 0.0705
Rheumatoid Arthritis,
cyclic citrullinated peptide (CCP) positive ENST00000318950 3 rs360136:C>A 0.0226 0.0651
Multiple complex diseases ENST00000361905 3 rs16911839:C>G 0.0188 0.0363
Amyloid A Levels ENST00000396253 3 rs12416821:T>C 0.0472 0.0243
Type II Diabetes Mellitus ENST00000327470 3 rs10500938:C>T 0.3011 0.0968
Urinary metabolites ENST00000321505 3 rs7123662:A>T 0.0837 0.0352
GWAS of log10 serum total
immunoglobulin E concentration
in a British population
ENST00000353172 3 rs3824865:A>G 0.0716 0.0923
Multiple complex diseases ENST00000279441 3 rs470171:C>G 0.0841 0.0443
Cardiovascular disease risk factors ENST00000530849 3 rs1047964:G>C 0.1902 0.0221
Multiple complex diseases ENST00000357529 3 rs9783460:C>A 0.3374 0.0609
Alpha-2-Macroglobulin ENST00000318602 5 rs226380:T>G 0.0111 0.0445
Multiple continuous traits
in DGI samples ENST00000486433 3 rs2638315:C>G 0.0073 0.0597
Multiple continuous traits
in DGI samples ENST00000378485 3 rs17251627:A>G 0.2998 0.0355
Insulin resistance/response ENST00000360185 3 rs7314498:G>A 0.0363 0.0663
Type 1 diabetes;
Multiple complex diseases ENST00000550722 3 rs3519:C>T 0.2340 0.0653
Suicide attempts in bipolar disorder ENST00000382298 5 rs17078720:A>G 0.3550 0.0404
Parkinson‘s disease ENST00000554271 3 rs7560:T>G 0.1258 0.0516
Multiple complex diseases ENST00000335725 3 rs9488:C>T 0.0698 0.0283
Breast Neoplasms ENST00000396402 3 rs4646:T>G 0.2089 0.0501
Smoking behavior;Lung cancer;
Chronic obstructive pulmonary
disease;Lung adenocarcinoma
ENST00000258886 3 rs1062980:T>C 0.0270 0.0121
Smoking behavior; ENST00000044462 5 rs3813570:T>C 0.0043 0.0542
Lung adenocarcinoma ENST00000261751 3 rs1948:T>C 0.0888 0.0615
Longevity ENST00000284382 3 rs12914235:A>G 0.2592 0.0202
Insulin-like growth factors ENST00000262302 3 rs1065656:C>G 0.0116 0.0687
Insulin resistance/response ENST00000318282 3 rs30126:C>T 0.1501 0.0382
Multiple complex diseases ENST00000540146 3 rs1054028:T>C 0.0105 0.0231
Hemoglobin A, Glycosylated ENST00000324015 3 rs1057355:G>T 0.1312 0.0889
Allergic rhinitis ENST00000322957 3 rs3192453:G>C 0.2871 0.0622
Myopia (pathological) ENST00000306329 3 rs3744975:C>T 0.0033 0.0194
15Sabarinathan et al., Human Mutation
Table S4 - continued
Ensembl p-value
Disease/phenotype id UTR dbSNP dmax/RNAfold rmin/RNAfold
Bladder cancer ENST00000436407 5 rs10432193:T>C 0.0618 0.0533
Multiple continuous traits
in DGI samples ENST00000334889 3 rs9947104:T>C 0.0199 0.0908
Gallstone disease ENST00000251047 3 rs1043334:A>C 0.2259 0.0403
Alcohol dependence ENST00000302850 3 rs1864193:G>T 0.1657 0.0978
Suicide attempts in
bipolar disorder ENST00000222249 3 rs1057261:A>G 0.1008 0.0751
Urinary Tract Infections;
Vesico-Ureteral Reflux
ENST00000243578 3 rs1800468:G>A 0.1071 0.0814
Alcohol dependence; ENST00000300843 3 rs344797:T>G 0.0037 0.0745
Multiple continuous traits
in DGI samples ENST00000262919 3 rs432647:T>C 0.1697 0.0284
Suicide attempts in
bipolar disorder
ENST00000246006 3 rs7492:T>C 0.0380 0.0311
Plasma levels of Protein C ENST00000246186 3 rs6060341:A>G 0.3550 0.0963
Parkinson‘s disease; ENST00000244061 3 rs6125829:G>T 0.0446 0.0674
GWAS of bipolar disorder
in the Japanese population ENST00000284987 3 rs229070:C>G 0.3589 0.0856
16Sabarinathan et al., Human Mutation
References
[1] You S, Stump DD, Branch AD, Rice CM. 2004. A cis-acting replication element in the sequence encoding
the NS5B RNA-dependent RNA polymerase is required for hepatitis C virus RNA replication. J Virol
78:1352-1366.
[2] Grover R, Sharathchandra A, Ponnuswamy A, Khan D, Das S. 2011. Effect of mutations on the p53
IRES RNA structure: Implications for de-regulation of the synthesis of p53 isoforms. RNA Biol 8:132-
142.
[3] Westerhout EM, Ooms M, Vink M, Das AT, Berkhout B. 2005. HIV-1 can escape from RNA interference
by evolving an alternative structure in its RNA genome. Nucleic Acids Res 33:796-804.
[4] Shen LX, Basilion JP, Stanton VP. 1999. Single-nucleotide polymorphisms can cause different structural
folds of mRNA. Proc Natl Acad Sci U S A, 96:7871-7876.
[5] Teresi RE, Zbuk KM, Pezzolesi MG, Waite KA, Eng C. 2007. Cowden syndrome-affected patients with
PTEN promoter mutations demonstrate abnormal protein translation. Am J Hum Genet, 81:756-767.
[6] van Hemert FJ, Zaaijer HL, Berkhout B, Lukashov VV. 2008. Occult hepatitis B infection: an evolu-
tionary scenario. Virology J 5:146.
[7] Chen GL, Vallender EJ, Miller GM. 2008. Functional characterization of the human TPH2 5’ regulatory
region: untranslated region and polymorphisms modulate gene expression in vitro. Hum Genet 122:645-
657.
[8] Carpen JD, Archer SN, Skene DJ, Smits M, von Schantz M. 2005. A single-nucleotide polymorphism
in the 5’-untranslated region of the hPER2 gene is associated with diurnal preference. J Sleep Res
14:293-297.
[9] Song C, Chen LZ, Zhang RH, Yu XJ, Zeng YX. 2006. Functional variant in the 3’-untranslated region
of Toll-like receptor 4 is associated with nasopharyngeal carcinoma risk. Cancer Biol Ther, 5:1285-1291.
[10] Nury D, Chabanon H, Levadoux-Martin M, Hesketh J.. 2005. An eleven nucleotide section of the 3’-
untranslated region is required for perinuclear localization of rat metallothionein-1 mRNA. Biochem J
387:419-428.
[11] Piri N, Gao YQ, Danciger M, Mendoza E, Fishman GA, Farber DB. 2005. A substitution of G to C in
the cone cGMP-phosphodiesterase gamma subunit gene found in a distinctive form of cone dystrophy.
Ophthalmology, 112:159-166.
[12] Reamon-Buettner SM, Cho SH, Borlak J. 2007. Mutations in the 3’-untranslated region of GATA4 as
molecular hotspots for congenital heart disease (CHD). BMC Med Genet 8:38.
[13] Hill MJ, Reynolds GP. 2011. Functional consequences of two HTR2C polymorphisms associated with
antipsychotic-induced weight gain. Pharmacogenomics, 12:727-734.
[14] Siala O, Salem IH, Tlili A, Ammar I, Belguith H, Fakhfakh F. 2010. Novel sequence variations in
LAMA2 and SGCG genes modulating cis-acting regulatory elements and RNA secondary structure.
Genet Mol Bio 33:190-197.
[15] Tsao D, Shabalina SA, Gauthier J, Dokholyan NV, Diatchenko L. 2011. Disruptive mRNA folding
increases translational efficiency of catechol-O-methyltransferase variant. Nucleic Acids Res 39:6201-
6212.
17Sabarinathan et al., Human Mutation
[16] Boffa MB, Maret D, Hamill JD, Bastajian N, Crainich P, Jenny NS, Tang Z, Macy EM, Tracy RP, Franco
RF, Nesheim ME, Koschinsky ML. 2008. Effect of single nucleotide polymorphisms on expression of the
gene encoding thrombin-activatable fibrinolysis inhibitor: a functional analysis. Blood, 111:183-189.
[17] Thiel CT, Horn D, Zabel B, Ekici AB, Salinas K, Gebhart E, Rschendorf F, Sticht H, Spranger J,
Mller D, Zweier C, Schmitt ME, Reis A, Rauch A. 2005. Severely incapacitating mutations in patients
with extreme short stature identify RNA-processing endoribonuclease RMRP as an essential cell growth
regulator. Am J Hum Genet 77:795-806.
[18] Venturin M, Moncini S, Villa V, Russo S, Bonati MT, Larizza L, Riva P. 2006. Mutations and novel
polymorphisms in coding regions and UTRs of CDK5R1 and OMG genes in patients with non-syndromic
mental retardation. Neurogenetics, 7:59-66.
[19] Tang S, Collier AJ, Elliott RM. 1999. Alterations to both the primary and predicted secondary structure
of stem-loop IIIc of the hepatitis C virus 1b 5’ untranslated region (5’UTR) lead to mutants severely
defective in translation which cannot be complemented in trans by the wild-type 5’UTR sequence. J
Virol 73:2359-2364.
[20] Camaschella C, Zecchina G, Lockitch G, Roetto A, Campanella A, Arosio P, Levi S. 2000. A new
mutation (G51C) in the iron-responsive element (IRE) of L-ferritin associated with hyperferritinaemia-
cataract syndrome decreases the binding affinity of the mutated IRE for iron-regulatory proteins. Br J
Haematol 108:480-482.
[21] Griseri P, Lantieri F, Puppo F, Bachetti T, Di Duca M, Ravazzolo R, Ceccherini I. 2007 A common
variant located in the 3’UTR of the RET gene is associated with protection from Hirschsprung disease.
Hum Mutat 28:168-76.
[22] Tang Z, Diamond MA, Chen JM, Holly TA, Bonow RO, Dasgupta A, Hyslop T, Purzycki A, Wagner
J, McNamara DM, Kukulski T, Wos S, Velazquez EJ, Ardlie K, Feldman AM. 2007. Polymorphisms
in adenosine receptor genes are associated with infarct size in patients with ischemic cardiomyopathy.
Clin Pharmacol Ther 82:435-440.
[23] Vallender EJ, Priddy CM, Hakim S, Yang H, Chen GL, Miller GM. 2008. Functional variation in the
3’ untranslated region of the serotonin transporter in human and rhesus macaque. Genes Brain Behav
7:690-697.
[24] Capasso M, Ayala F, Russo R, Avvisati RA, Asci R, Iolascon A. 2009. A predicted functional single-
nucleotide polymorphism of bone morphogenetic protein-4 gene affects mrna expression and shows a
significant association with cutaneous melanoma in southern italian population. J Cancer Res Clin
Oncol 135:1799-1807.
18