G Model

FSIGEN-790; No. of Pages 5

Forensic Science International: Genetics xxx (2011) xxx–xxx

Contents lists available at SciVerse ScienceDirect

Forensic Science International: Genetics

journal homepage: www.elsevier.com/locate/fsig

Short communication

A model for data analysis of microRNA expression in forensic body fluid

identification
Zheng Wang a, Haibo Luo a, Xiongfei Pan b, Miao Liao a, Yiping Hou a,*
a
Department of Forensic Genetics, West China School of Basic Science and Forensic Medicine, Sichuan University (West China University of Medical Sciences), Chengdu 610041,
Sichuan, China
b
Department of Epidemiology, West China School of Public Health, Sichuan University, Chengdu 610041, Sichuan, China

A R T I C L E I N F O A B S T R A C T

Article history: MicroRNAs (miRNAs, 18–25 bases in length) are small, non-coding RNAs that regulate gene expression at
Received 30 July 2010 the post-transcriptional level. MiRNA expression patterns, including presence and relative abundance of
Received in revised form 9 August 2011 particular miRNA species, provide cell- and tissue-specific information that can be used for body fluid
Accepted 18 August 2011
identification. Recently, two published studies reported that a number of body fluid-specific miRNAs had
been identified. However, the results were inconsistent when different technology platforms and
Keywords: statistical methods were applied. To further study the role of miRNAs in identification of body fluids, this
miRNA
study sets out to develop an accurate and reliable model for data analysis of miRNA expression. To that
Body fluid identification
TaqMan
end, the relative expression levels of three miRNAs were studied using the mirVanaTM miRNA Isolation
RT-qPCR Kit, high-specificity stem-loop reverse transcription (RT) and high-sensitivity hydrolysis probes
Forensic medicine (TaqMan) quantitative real-time polymerase chain reaction (qPCR) in forensically relevant biological
fluids, including venous blood, vaginal secretions, menstrual blood, semen and saliva. Accurate
quantification of miRNAs requires not only a highly sensitive and specific detection platform for
experiment operation, but also a reproducible methodology with an adequate model for data analysis. In
our study, the efficiency-calibrated model that incorporated the impact of the quantification cycle (Cq)
values and PCR efficiencies of target and reference genes was developed to calculate the relative
expression ratio of miRNAs in forensically relevant body fluids. Our results showed that venous blood
was distinguished from other body fluids according to the relative expression ratio of miR16 using as
little as 50 pg of total RNA, while the expression level of miR658 was unstable and that of miR205 was
nonspecific among different body fluids. Collectively, the findings may constitute a basis for future
miRNA-based research on body fluid identification and show miRNAs as a promising biomarker in
forensic identification of body fluids.
ß 2011 Elsevier Ireland Ltd. All rights reserved.

1. Introduction in a tissue-specific manner and their expression patterns can

In forensic casework, especially in cases of sexual assault and
child sexual abuse, to uncover the criminal nature of an event
requires not only identification of the DNA profile, but also
confirmation of the origin of DNA. One of the challenges facing the
forensic community is how to find a reliable biomarker and
develop an accurate method for rapid identification of body fluids.
Conventional serology-based methods for body fluid identifica-
tion are prone to various limitations, such as sample consump-
tion, intensive labor, time consumption, varying degrees of
sensitivity and specificity, and no definitive tests for the presence
of menstrual blood and vaginal secretions [1–5]. Numerous
published studies have reported that some mRNAs are expressed
* Corresponding author. Tel.: +86 28 85501550; fax: +86 28 85501549.
E-mail address: prof.hou@yahoo.cn (Y. Hou).
confirm specific body fluids, even after long periods under
controlled conditions [6–8]. However, heat, humidity, UV light
and ubiquitous ribonucleases are detrimental to mRNA stability
as a sensitive and specific biomarker for forensic applications
[9,10].
MiRNAs belong to a class of small, non-coding RNA molecules
that regulate gene expression at the post-transcriptional level [11–
13]. Numerous published studies have demonstrated their
important role in functions including embryonic development,
proliferation, hematopoiesis and apoptosis, and in the pathogene-
sis of many human diseases such as viral infection, cancer and
metabolic disorders [12–14]. Several studies revealed that many
miRNAs were expressed in a tissue-specific manner [15,16]. The
intrinsically short fragment and tissue-specific expression enable
miRNA as an ideal biomarker for body fluid identification. Recently,
two published studies reported their preliminary application in
forensic identification of body fluids [17,18]. However, the results

1872-4973/$ – see front matter ß 2011 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.fsigen.2011.08.008

Please cite this article in press as: Z. Wang, et al., A model for data analysis of microRNA expression in forensic body fluid identification,
Forensic Sci. Int. Genet. (2011), doi:10.1016/j.fsigen.2011.08.008
G Model
FSIGEN-790; No. of Pages 5
2 Z. Wang et al. / Forensic Science International: Genetics xxx (2011) xxx–xxx
were inconsistent when different technology platforms and
statistical methods were applied.
In this study, we used the mirVanaTM miRNA Isolation Kit that
optimized extraction of small nucleic acid molecules to enrich
miRNAs. The expression abundance was detected using stem-loop
reverse transcription (RT) followed by TaqMan quantitative real-
time polymerase chain reaction (qPCR) analysis. We also utilized
the gradient dilution cDNA method to test the amplification
efficiencies of target and reference genes in body fluids [19–21].
The efficiency-calibrated method was used to calculate the relative
expression ratio of target genes in body fluids [19]. To further
explore its potential application in forensic cases, we used serial
dilutions of total RNA for cDNA synthesis to establish the detection
sensitivity of TaqMan RT-PCR assays in venous blood.
2. Materials and methods
2.1. Collection of body fluid samples
The body fluid samples were collected on sterile cotton swabs
and dried at room temperature. Blood samples were collected by
venipuncture without anticoagulation treatment and 50 ml
aliquots were spotted onto sterile cotton swabs. Semen-free
vaginal secretions and menstrual blood were collected from the
vagina on sterile cotton swabs and dried at room temperature.
Freshly ejaculated semen samples were provided in sealed plastic
cups and dried onto sterile cotton swabs. Saliva samples were
provided in sealed plastic tubes and dried onto sterile cotton
swabs. Buccal samples were collected from donors using sterile
swabs by swabbing the inside of mouth. Each body fluid sample
was collected from 10 unrelated individuals (age: 18–47 years old)
of the Chinese Han population living in Sichuan province. Written
informed consent was obtained from all individuals. All the
samples were stored at 80 8C for future use. A single cotton swab
was used for RNA isolation.
2.2. RNA isolation and quantification
RNA was isolated using the mirVanaTM miRNA Isolation Kit
(Ambion, Austin, TX, USA) according to the manufacturer’s
instructions. The purity and quantity of RNA were assessed using
the NanoDrop ND-1000 spectrophotometer (Thermo Scientific,
Wilmington, DE, USA). All the samples were diluted to a final
concentration of 10 ng/ml. The samples were used immediately or
stored at 80 8C for future use.
2.3. TaqMan RT-qPCR
The TaqMan1 MicroRNA Reverse Transcription Kit (Applied
Biosystems, Foster City, CA, USA) was used for preparation of cDNA.
RT reactions were performed in a volume of 15 ml, and each
reaction contained 10 ng of total RNA. RT reactions were
performed on a GeneAmpPCR System 9600 (Applied Biosystems)
with the following conditions: 16 8C for 30 min, 42 8C for 30 min,
85 8C for 5 min, and 4 8C on hold. Reactions without addition of
reverse transcriptase (RT () controls) were performed alongside
with cDNA synthesis of each sample and used in subsequent
procedures to control the potential genomic DNA contamination.
1 ml of RT reaction product was added in the 20 ml qPCR reaction.
All TaqMan assays were run in triplicate on an ABI Prism 7500
using TaqMan1 Universal PCR Master Mix II without UNG (Applied
Biosystems). Real-time PCR cycling conditions consisted of 95 8C
for 10 min, followed by 50 cycles of 95 8C for 15 s and 60 8C for
1 min. The Cq is defined as the PCR cycle at which the fluorescent
signal of the reporter dye crosses an arbitrarily placed threshold in
the exponential phase. For our assays, the constant threshold value
Please cite this article in press as: Z. Wang, et al., A model for data analysis of microRNA expression in forensic body fluid identification,
Forensic Sci. Int. Genet. (2011), doi:10.1016/j.fsigen.2011.08.008
was set to 0.200 for easy comparison of results within and between
assays. The sample with a Cq value over 40 or an amplification
point not reaching the threshold was considered as invalid and was
not useful for further analysis.
2.4. Data analysis
The three miRNAs assay was controlled with the snRNA (U6b)
assay (Applied Biosystems, Supplemental Table 1) functioning as
reference gene to evaluate Cq values and serving as an internal
positive control if sample extraction and reverse transcription
were performed successfully.
The relative expression ratio (R) is expressed as the target/
reference ratio of each sample normalized by the target/reference
ratio of the normalizer [19,22].
ðEre fx þ 1ÞCqre fx ðEre f þ 1ÞCqre f
R¼ 
ðEtarx þ 1Þ Cqtarx
ðEtar þ 1ÞCqtar
The above equation shows the mathematical model of relative
expression ratio in qPCR. The ratio of a target gene is expressed in a
body fluid versus the normalizer (in this study, venous blood was
designated as the normalizer) in comparison to the reference gene
(U6b). Etarx is the real-time PCR efficiency of target gene in x body
fluids (x means vaginal secretions, menstrual blood, semen, saliva
or buccal swabs); Erefx is the real-time PCR efficiency of reference
gene in x body fluid; Cqtarx is the Cq value of target gene in x body
fluid; Cqrefx is the Cq value of reference gene in x body fluid; Etar is
the real-time PCR efficiency of target gene in venous blood; Eref is
the real-time PCR efficiency of reference gene in venous blood;
Cqtar is the Cq value of target gene in venous blood; and Cqref is the
Cq value of reference gene in venous blood.
For calculation of R, the qPCR efficiencies and Cq values of
investigated transcripts must be known. Serial dilutions of cDNA
over a 100-fold range were prepared. According to linear
regression equation and E = 10[1/slope]-1, the amplification effi-
ciencies were calculated. R > 1 indicates higher abundance of the
target gene in x body fluid than venous blood, while R < 1 suggests
otherwise.
2.5. Analytical sensitivity of miRNA TaqMan assays
We used serial dilutions of total RNA (10 ng to 0.01 ng) isolated
from venous blood as input for cDNA synthesis to establish the
detection sensitivity of TaqMan RT-PCR assays.
3. Results
Relative abundance of three miRNAs relative to U6b among
body fluids were presented and average Cq values of ten unrelated
individuals were tabulated (Supplemental Table 2). The expression
abundance of miR658 was lower than U6b in all body fluid
samples. MiR658 was unstably expressed in the samples,
especially in the venous blood (not detectable in 7 of 10 blood
samples), and apparent inter-individual variation was observed in
vaginal secretions. The expression of miR205 had much higher
abundance in buccal swabs, vaginal secretions and menstrual
blood compared with saliva, semen and venous blood, but it could
not be distinguished from those in vaginal secretions and
menstrual blood. The expression of miR16 varied significantly
between the venous blood and other body fluids (Supplemental
Fig. 1). Obvious expression difference (dCt6) was observed,
though the abundance of miR16 was high in both the menstrual
and venous blood. The RT () controls and no template controls
were negative in all cases. Therefore, miR16 as a potential body
fluid-specific miRNA was selected for further study.
G Model
FSIGEN-790; No. of Pages 5

Z. Wang et al. / Forensic Science International: Genetics xxx (2011) xxx–xxx 3

Fig. 1. Determination of qPCR efficiencies from the slopes of the calibration curve, according to the equation: E = 10[1/slope]-1of miR16 and U6b. Cq values versus cDNA
concentration input (log scale) were plotted to calculate the slope (mean  SD; n = 3).

The Cq values of miR16 were plotted versus the log of the
dilution cDNA and the linear regressions were performed from
which the mean efficiency could be derived (Fig. 1). The relative
expression ratios of miR16 in different body fluids were compared
(Fig. 2).
In sensitivity testing, miR16 and U6b were detectable in the
target body fluid using an amount as low as 50 pg total RNA for
cDNA synthesis (Supplemental Fig. 2).
4. Discussion
Since the discovery of miRNAs [23], many researchers have
explored their tissue-specific expression patterns [15,16]. Recent-
ly, two published studies reported their preliminary application in
forensic identification of body fluids [17,18]. Hanson et al. [17]
identified 9 body fluid-specific miRNAs using random and oligo-dT
primers for RT and SYBR Green for qPCR, whereas Zubakov et al.

Please cite this article in press as: Z. Wang, et al., A model for data analysis of microRNA expression in forensic body fluid identification,
Forensic Sci. Int. Genet. (2011), doi:10.1016/j.fsigen.2011.08.008
G Model
FSIGEN-790; No. of Pages 5
4 Z. Wang et al. / Forensic Science International: Genetics xxx (2011) xxx–xxx
Fig. 2. Relative expression ratio of miR16 in body fluids, i.e., venous blood (VB),
vaginal secretions (VS), menstrual blood (MB), semen (SE), saliva (SA) and buccal
swabs (BS). Values underwent log 10 transformation.
[18] adopted the stem-loop RT primers for RT and TaqMan probe
for qPCR in tracing another 14 body fluid-specific miRNAs. Body
fluid-specificity of miRNAs identified by the two groups were
inconsistent and Zubakov et al. claimed that the specific markers in
vaginal secretions, menstrual blood and saliva suggested by
Hanson et al. could not be replicated through their own methods.
One possible explanation is that they used different technology
platforms, which implies different strategies for both cDNA
synthesis and PCR quantification.
Accurate quantification of miRNAs requires not only a high
sensitive and specific detection platform for experiment operation,
but also a reproducible methodology with an adequate model for
data analysis.
The preparation of total RNA or pure miRNAs is the prerequisite
for accurate quantification. The amount of miRNAs in a total RNA
sample depends on the recovery efficiency that may be signifi-
cantly affected by the different purification methods employed
[24]. In this study, we used the mirVanaTM miRNA Isolation Kit that
is designed for purification of RNA suitable for studies of both
siRNA and miRNA in natural populations.
The miRNA expression can be detected and quantified by
several methods including northern blotting, microarrays, and
real-time PCR. Northern blotting cannot discriminate mature and
precursor miRNAs (pre-miRNA) and poses potential risks to
researchers exposed to certain chemicals [25]. Though microarrays
could improve the throughput of miRNA profiling, they are
subjected to low sensitivity and specificity, or high expense
[26]. The qPCR has become a gold standard of nucleic acid
quantification due to the specificity and sensitivity [19–21].
Detection chemistries for qPCR include SYBR Green and TaqMan.
Since SYBR Green binds nonspecifically to all double-stranded DNA
sequences, melt curve or gel analysis is required to detect
nonspecific products in case of false-positive signals. In our
research we used the TaqMan MicroRNA assay, which included
stem-loop RT primers for RT and TaqMan MGB (minor groove
binder) probes for qPCR. Due to the base stacking and spatial
constraint of the stem–loop structure, stem–loop RT primers are
advantageous over conventional linear ones in terms of RT
efficiency and specificity [27]. The TaqMan MGB probes contain
a reporter dye (FAMTM dye) linked to the 50 end of the probe and a
nonfluorescent quencher (NFQ) with a MGB at the 30 end of the
probe. This MGB modification increases the melting temperature
(Tm) without increasing probe length, which allows the design of
shorter probes [27]. During PCR, the DNA polymerase cleaves only
probes that are specifically hybridized to the target. Cleavage
Please cite this article in press as: Z. Wang, et al., A model for data analysis of microRNA expression in forensic body fluid identification,
Forensic Sci. Int. Genet. (2011), doi:10.1016/j.fsigen.2011.08.008
separates the reporter dye from the quencher dye and leads to
increased fluorescence by the reporter.
The accuracy and success of qPCR analysis depend on proper
normalization of data. The purpose of normalization is to minimize
potential variation that can mask or exaggerate biologically
meaningful changes. Although Vandesompele et al. [28] proposed
to use multiple reference genes as a normalization factor for data
analysis, the study of miRNAs in forensically relevant body fluids is
at an exploratory stage and there are no generally accepted
multiple reference genes. U6b was chosen based on its high
abundance (P/N: 044972, Applied Biosystems) and apparent
stability in different body stains of forensic interest [17].
How to present the data is an essential issue in gene expression
profiling using qPCR. Several methods exist to analyze the relative
gene expression. One simple method to process qPCR, currently
known as the DDCq method, is based solely on Cq values [29].
However, this method assumes that all amplification efficiencies
are equal to 1. Therefore it does not take into consideration
possible variations of amplification efficiencies. Recently, Mest-
dagh et al. [30] have described an alternative approach to
normalize qPCR experiments, using the mean miRNA expression
value as normalization method. Although this method generates
stable results, which performs better than using snoRNAs for
normalization, it is applicable only when adequate miRNAs are
analyzed [31]. Soong et al. [22] proposed the efficiency-calibrated
mathematical method for the relative expression ratio in qPCR. The
equation is, in principle, the same and the results are identical to
those using the DDCq method under the condition of Etar = Eref = 1.
The advantage of the efficiency-calibrated method is that the PCR
efficiencies of the target and reference genes are included in the
equation, making the data analysis more accurate.
MiRNAs in body fluids can be expressed in three different
patterns: differential expression, which can be detectable to
distinguish body fluids; universal expression, which can be
potentially considered as the internal reference in the future
(related research in progress in our lab); and unstable expression,
which may be affected by the physiological state. MiRNAs with
these three expression patterns were selected to analyze the
expression abundance in different body fluids using the efficiency-
calibrated model in this study. This research demonstrated that
miR16 was venous blood-specific and was detectable using as little
as 50 pg of total RNA. However, the expressed miR658 and miR205
were not saliva-specific, which is inconsistent with the result from
Hanson et al. [17], and their expression in saliva and buccal swabs
differed. Unstable expression of miR658 in body fluids may be
affected by physiological conditions, and miR205 may be
epithelium-specific so that other miRNAs should be considered
to distinguish vaginal and oral epithelial cells.
Different tissues exhibit different PCR efficiencies, caused by RT
inhibitors, PCR inhibitors, and variations in the total RNA fraction
pattern extracted. Numerous published studies reported that the
PCR efficiency has a major impact on the accuracy of the calculated
expression result [19–21,32]. Difference in PCR efficiency (DE) of
3% (DE = 0.03) between target gene and reference gene generate a
falsely calculated differences in expression ratio of 47% in case of
Etarget < Eref and 209% in case of Etarget > Eref after 25 performed
cycles [32]. We calculated the amplification efficiencies of miRNAs
in different body fluids using the gradient dilution cDNA method
and the relative expression ratio based on the efficiency-calibrated
model. We found that the calculation without efficiency correction
could generate false differences in expression ratios of 30%
(menstrual blood), 17% (semen), 6% (saliva), 6% (vaginal secretions)
and 3% (buccal swabs), which indicates that the efficiency-
calibrated model analysis is more accurate.
In conclusion, this article presents a simple procedure for
miRNA-based body fluid identification and an accurate model for
G Model
FSIGEN-790; No. of Pages 5
Z. Wang et al. / Forensic Science International: Genetics xxx (2011) xxx–xxx
data analysis. Forensic biomarkers for body fluid identification
should have not only high expression abundance, but also constant
expression levels in individuals. Currently, over 1000 mature
miRNAs have been identified in the human genome (miRBase v16,
September 2010). It remains necessary to search for suitable
miRNA markers for forensically relevant body fluids. More
research should be dedicated towards identification of optimum
miRNAs for body fluid identification and analysis of potential
influences of environmental factors such as humidity, UV radiation
and bacterial contamination on miRNA stability in vitro.
Conflict of interest
None.
Acknowledgement
This study was supported by grants (nos. 81072510 and
30872917) from the National Nature Science Foundation, China.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.fsigen.2011.08.008.
References
[1] R.E. Gaensslen, Sourcebook in Forensic Serology, Immunology, and Biochemistry,
U.S. Department of Justice, Washington, DC, 1983.
[2] A.C. Ponce, F.A.V. Pascual, Critical revision of presumptive tests for bloodstains,
Forensic Sci. Commun. 1 (2) (1999) 1–15.
[3] N. Khaldi, A. Miras, K. Botti, L. Benali, S. Gromb, Evaluation of three rapid detection
methods for the forensic identification of seminal fluid in rape cases, J. Forensic
Sci. 49 (2004) 749–753.
[4] S.S. Tobe, N. Watson, N.N. Daéid, Evaluation of six presumptive tests for blood,
their specificity, sensitivity, and effect on high molecular-weight DNA, J. Forensic
Sci. 52 (2007) 102–109.
[5] J.R. Mayers, W.K. Adkins, Comparison of modern techniques for saliva screening, J.
Forensic Sci. 53 (2008) 862–867.
[6] J. Juusola, J. Ballantyne, Messenger RNA profiling: a prototype method to supplant
conventional methods for body fluid identification, Forensic Sci. Int. 135 (2003)
85–96.
[7] J. Juusola, J. Ballantyne, Multiplex mRNA profiling for the identification of body
fluids, Forensic Sci. Int. 152 (2005) 1–12.
[8] J. Juusola, Ballantyne, mRNA profiling for body fluid identification by multiplex
quantitative RT-PCR, J. Forensic Sci. 52 (2007) 1252–1262.
[9] M. Setzer, J. Juusola, J. Ballantyne, Recovery and stability of RNA in vaginal swabs
and blood, semen, and saliva stains, J. Forensic Sci. 53 (2008) 296–305.
[10] D. Zubakov, M. Kokshoorn, A. Kloosterman, M. Kayser, New markers for old stains:
stable mRNA markers for blood and saliva identification from up to 16-year-old
stains, Int. J. Legal Med. 123 (2009) 71–74.
Please cite this article in press as: Z. Wang, et al., A model for data analysis of microRNA expression in forensic body fluid identification,
Forensic Sci. Int. Genet. (2011), doi:10.1016/j.fsigen.2011.08.008
5
[11] D.P. Bartel, MicroRNAs: genomics, biogenesis, mechanism, and function, Cell 116
(2004) 281–297.
[12] H.W. Hwang, J.T. Mendell, MicroRNAs in cell proliferation, cell death, and tumor-
igenesis, Br. J. Cancer 94 (2006) 776–780.
[13] T.M. Rana, Illuminating the silence: understanding the structure and function of
small RNAs, Nat. Rev. Mol. Cell Biol. 8 (2007) 23–36.
[14] S.F. Tavazoie, C. Alarcón, T. Oskarsson, D. Padua, Q. Wang, P.D. Bos, W.L. Gerald, J.
Massagué, Endogenous human microRNAs that suppress breast cancer metasta-
sis, Nature 451 (2008) 147–152.
[15] P. Sood, A. Krek, M. Zavolan, G. Macino, N. Rajewsky, Cell-type-specific signatures
of microRNAs on target mRNA expression, Proc. Natl Acad. Sci. U. S. A. 103 (2006)
2746–2751.
[16] Y. Liang, D. Ridzon, L. Wong, C. Chen, Characterization of microRNA expression
profiles in normal human tissues, BMC Genomics 8 (2007) 166.
[17] E.K. Hanson, H. Lubenow, J. Ballantyne, Identification of forensically relevant body
fluids using a panel of differentially expressed microRNAs, Anal. Biochem. 387
(2009) 303–314.
[18] D. Zubakov, A.W. Boersma, Y. Choi, P.F.V. Kuijk, E.A. Wiemer, M. Kayser,
MicroRNA markers for forensic body fluid identification obtained from micro-
array screening and quantitative RT-PCR confirmation, Int. J. Legal Med. 124
(2010) 217–226.
[19] M.W. Pfaffl, A new mathematical model for relative quantification in real-time
RT-PCR, Nucleic Acids Res. 29 (2001) e45.
[20] Y. Karlen, A. McNair, S. Perseguers, C. Mazza, N. Mermod, Statistical significance of
quantitative PCR, BMC Bioinformatics 20 (2007) 131.
[21] J. Huggett, K. Dheda, S. Bustin, A. Zumla, Real-time RT-PCR normalisation;
strategies and considerations, Genes Immunity 4 (2005) 279–284.
[22] R. Soong, J. Ruschoff, K. Tabiti, Detection of Colorectal Micrometastasis by
Quantitative RT-PCR of Cytokeratin 20 mRNA, Roche Diagnostics Internal Publi-
cation, 2000.
[23] R.C. Lee, R.L. Feinbaum, V. Ambros, The C. elegans heterochronic gene lin-4
encodes small RNAs with antisense complementarity to lin-14, Cell 75 (1993)
843–854.
[24] A. Masotti, V. Caputo, L.D. Sacco, A. Pizzuti, B. Dallapiccola, G.F. Bottazzo, Quan-
tification of small non-coding RNAs allows an accurate comparison of miRNA
expression profiles, J. Biomed. Biotechnol. (2009) 659028.
[25] S. Streit, C.W. Michalski, M. Erkan, J. Kleeff, H. Friess, Northern blot analysis for
detection and quantification of RNA in pancreatic cancer cells and tissues, Nat.
Protoc. 4 (2009) 37–43.
[26] F. Sato, S. Tsuchiya, K. Terasawa, G. Tsujimoto, Intra-platform repeatability and
inter-platform comparability of microRNA microarray technology, PLoS One 5
(2009) e5540.
[27] C. Chen, D.A. Ridzon, A.J. Broomer, Z. Zhou, D.H. Lee, J.T. Nguyen, M. Barbisin, N.L.
Xu, V.R. Mahuvakar, M.R. Andersen, K.Q. Lao, K.J. Livak, K.J. Guegler, Real-time
quantification of microRNAs by stem-loop RT-PCR, Nucleic Acids Res. 33 (2005)
e179.
[28] J. Vandesompele, K.D. Preter, F. Pattyn, B. Poppe, N.V. Roy, A.D. Paepe, F. Speleman,
Accurate normalization of real-time quantitative RT-PCR data by geometric
averaging of multiple internal control genes, Genome Biol. 3 (2002) 34.
[29] K.J. Livak, T.D. Schmittgen, Analysis of relative gene expression data using real-
time quantitative PCR and the 2(-Delta Delta C(T)) method, Methods 25 (2001)
402–408.
[30] P. Mestdagh, P.V. Vlierberghe, A.D. Weer, D. Muth, F. Westermann, F. Speleman, J.
Vandesompele, A novel and universal method for microRNA RT-qPCR data
normalization, Genome Biol. 10 (2009) R64.
[31] V. Benes, M. Castoldi, Expression profiling of microRNA using real-time
quantitative PCR, how to use it and what is available, Methods 50 (2010)
244–249.
[32] M.W. Pfaffl, Quantification strategies in real-time PCR, in: S.A. Bustin (Ed.),
A–Z of Quantitative PCR, International University Line (IUL), La Jolla, 2004 ,
pp. 87–112.