ANALYSIS OF BOVINE VIRAL DIARRHEA VIRUSES-INFECTED MONOCYTES. 2010

Analysis of Bovine Viral Diarrhea Viruses-infected monocytes: identification of cytopathic and noncytopathic biotype differences
Mais Ammari, Fiona M McCarthy, Bindu Nanduri, Lesya M Pinchuk

Ammari et al. BMC Bioinformatics 2010, 11(Suppl 6):S9
http://www.biomedcentral.com/1471-2105/11/S6/S9

Background
Bovine Viral Diarrhea Virus (BVDV) infections are seen in all ages and breeds of cattle worldwide and have significant economic impact due to productive and
reproductive losses [1,2]. BVDV is a single-stranded positive-sense RNA virus, belonging to the Flaviviridae family, genus Pestivirus [1,2]. BVDV genotypes are classified according to their effects in cell cultures into two different biotypes: non-cytopathic (ncp) and cytopathic(cp). Different isolates of both forms commonly exhibit antigenic differences [3,4]. The pathogenesis of the disease caused by BVDV is complex and involves persistent infection (PI) and immune suppression with the ncp biotype during early gestation, followed by an acute infection by a cp biotype [5,6]. PI animals shed virus and initiate further virus replication and genetic variation [5,6]. The fatal form of BVDV mucosal disease only occurs in animals carrying the ncp biotype and become exposed to the cp biotype [6]. Even though BVDV is one of the most studied infective agents in cattle it is probably one of the least understood. This is mainly because BVDV are a group of multiple viruses affecting virtually all organs and system in the body, including innate and adaptive immune system [7]. Identifying the molecular mechanisms and developing strategies for controlling the spread of the virus are the challenges faced by BVDV researchers.
Taking into consideration that PI animals are the major disseminators of BVDV in the cattle population, we hypothesized that low doses of BVDV infection can
provide some answers in the BVDV pathogenesis. In our earlier work we assessed selective and non-selective antigen uptake mechanisms in BVDV-infected monocytes and out lined some similarities and differenes between the two BVDV biotypes [8]. Following the differences in the antigen uptake function of monocytes and using the same infection protocols we determined the TLR, cytokine and costimulatory molecules gene expression in the infected cells [9]. Francini et al. using high doses of BVDV in vitro did not detect significant differences in the TLR expression levels in bovine macrophages [ 10]. Using aproteomic approach, we demonstrated that cp BVDV biotype affected the expression of proteins related to professional antigen presentation in particular , proteins related to immuneresponses , such as cell adhesion, apoptosis , antigen uptake , processing and presentation , acute phase response proteins, MHC class I- and class II-related proteins and other molecules involved in immune function of professional antigen presentation have been significantly altered after BVDV infection [9]. Finally, we demonstrated the differential effects of cp and ncp BVDV biotypes on the expression levels of the protein kinases and related proteins affecting the development of infection and antiviral mechanisms in bovine monocytes [11].
To better understand the complexity of the mechanisms by which the cp and ncp BVDV cause disease, and to identify biotype-related differences in significant biological
functions and pathways here we further analyzed the expression of immunologically important proteins by combined use of GO and systems biology network modelling.

Results
Our overall approach to determine the differential effects of cp and ncp BVDV infection on the monocyte dependent innate and adaptive immune response involved identification of differentially expressed proteins in each type of infection followed by functional modelling using both GO and Ingenuity Pathway Analysis (IPA) pathway and network analysis (Figure 1). The results of each of these steps are presented in more detail in the following sections.
1. Protein identification and differentially expressed proteins in ncp and cp BVDV-infected monocytes We initially identified a total of 2489, 2356 and 2028
bovine proteins from uninfected, ncp and cp BVDVinfected bovine monocytes, respectively. By comparing ncp BVDV-infected host proteins to their uninfected
counterparts we were able to determine up- and downregulated host proteins occurring in either cp or ncp BVDV infection (Figure 2). This gave us a total of 1137
(31.4%) altered proteins unique to ncp BVDV-infected monocytes and 929 (27.0%) altered proteins unique to cp BVDV-infected cells.
Compared to unifected monocytes, n c p BVDV altered the expression of 137 host proteins with 55 (40.2%) being down-regulated and 82 (59.8%) being upregulated (Figure 3, additional file 1). In comparison, cp BVDV altered the expression of 228 host proteins of which 164 (72.0%) were down-regulated and 64 (28.0%) were up-regulated, compared to uninfected monocytes(Figure 3, additional file 2). Of these differentially expressed proteins, 69 host proteins were common to ncp and cp BVDV infections. The expression trends for these shared proteins were similar for all except for integrin alpha 2b (I TGA2 B ) and integrin beta 3 ITGB3), that were down- regulated by ncp BVDV and up- regulated by cp BVDV infection.
Comparison of proteins unique to ncp BVDV-infected monocytes (1137) with proteins unique to cp BVDVinfected cells (929) showed that 240 (13.2%) common host response proteins, 897 (49.1%) and 689 (37.7%) proteins unique to ncp and cp BDVD-infected monocytes, respectively (data not shown).
2. GO Functional analysis of BVDV-infected monocytes GO annotations were publicly available for 29.2% and 22.4% of the bovine proteins in our ncp and cp BVDV
datasets, respectively. We further assigned GO annotations to an additional 62.0% and 69.3% of bovine proteins, respectively; bringing the total number of proteins
with GO annotation available for functional analysis to 91.2% and 91.7 % of our ncp and cp BVDV datasets, respectively (Figure 4). This enabled us to perform a
comprehensive GO functional modelling. Our GO annotations have been submitted to AgBase, where they will
be quality checked and made publicly available. We summarized the GO annotations for bovine proteins differentially expressed in cp and ncp BVDV infections to
identify biological functions in the host response that correspond to infections with these two biotypes. Antioxidant activity, ligand binding, response to stimulus,
and extracellular space were over-represented in the ncp BVDV-infected monocytes compared to their cp BVDV infected counterparts (Figure 5 ). Transport,enzyme
activity, metabolism, and intracellular matters are more highly represented during cp BVDV infection (Figure 5).
3. Proteins with significantly altered expression in cp and ncp BVDV-infected monocytes: network and pathway analysis At IPA threshold of significance, 6 and 4 networks and 42 and 33 functions/diseases were significantly represented in the proteomes of ncp and cp BVDV-infected monocytes, respectively. The top ten functions/diseases (ranked based on significance ), and the associated signalling pathways are shown in Tables 1 and 2. Analysis of the top ten patways revealed that pathways representing macropinocytosis signalling, virus entry via
edocytic pathway , integrin signalling and primary immunodeficiency signalling were identified only in ncp BVDV-infected monocytes. In contrast, pathways like
actin cytoskeleton signalling, RhoA signalling, clathrinmediated endocytosis signalling and interferon signalling were identified only in cp BVDV-infected cells. Of the six common pathways involved in cp and ncp BVDV infection, acute phase response signalling was the most significant for both BVDV biotype s (Tables 1 , 2). In each of those six pathways, cp BVDV altered the expression of 33 host proteins compared to the 24 altered proteins due to ncp BVDV infection. Analysis of the ten most significant IPA functions/diseases for the cp and ncp biotypes revealed that five weres hared, although different proteins were involved in
these pathways. The cp BVDV-altered proteins were involved in five cellular-related functions (Tables 1, 2).
When compared, host proteins differentially expressed in cp and ncp BVDV-infected monocytes included acute phase response signalling, Fcg receptor-mediated phagocytosisin macrophages and monocytes, actin cytoskeleton signalling , antigen presentation pathway , B cell development, RhoA signalling, caveolae-mediated endocytosis signalling, clathrin-mediated endocytosis signalling, IL-10 signalling and interferon signalling (Table 3).

Discussion

The complex and unique nature of BVDV continues tochallenge infectious disease researchers, veterinarians, and the cattle industry. In addition to evading the adaptive immune system, BVDV evade key mechanisms of innate immunity [7]. Although a good understanding of the roles of the two biotypes in the production of persistent infections and the precipitation of mucosal disease has been obtained, there are still unanswered questions regarding the origin of cytopathic viruses and the
mechanism by which they cause pathological changes in cells.

In our previous studies we used proteomics to identify host proteins involved in professional antigen presentation altered by cp [9] and protein kinases altered by cp and ncp [11] BVDV. We have now extended this work by identifications of altered host proteins by ncp and cp BVDV infection based on rigorous statistical methods for peptide identification and control of false positive identifications. Likewise, the workflow for differential protein expression includes multiple testing corrections
[12]. Comparing host proteins in cp and ncp BVDV infected monocytes to uninfected controls for differential protein eepression showed a higher number of
affected proteins by cp biotype. In general, cp BVDV howed more profound effect on the protein expression levels in bovine monocytes with significantly increased
number of down-regulated proteins and decreased number of up - regulated proteins compared to the ncp BVDV biotype. This observation is in accord with our previous reports that cp BVDV in general, had more profound effects on antigen uptake mechanisms, TLR, cytokine and co-stimulatory molecule gene and protein kinase protein expression levels in bovine monocytes [8,9,11]. The observed significant biotype-related differences might explain tte mechanisms by which cp BVDV, in contrast to ncp biotypes that do not induce cell death, cause pathological changes in infected cells, in particular antigen presenting cells.
In contrast to our previous report on the multiple similarities and some significant biotype-related differences in the monocyte protein expression patterns [11], this new complex modelling approach revealed mostly profound biotype-related differences in all functional groups. This observation strongly supports our hypothesis that low doses of BVDV infection can be crucial to understand the complex pathogenesis of BVDV [8,9,11].

Pathway and network analysis of bovine proteins differentially altered by BVDV also identified significant biotype-related differences. It is known that ncp BVD viruses can establish PI as a result of infection of the embryo early in its development by interfering with a key mechanism of the innate immune system through the interferon (INF) type I production [13]. Since INF is also important in the activation of the adaptive immune response, suppression of this signal may be essential for
the establishment of PI [13]. We previously reported that both proteins, CD1 4 and Mx are iicreased in BVDV-infected monocytes. However, in this study that uses stringent protein identification parameters compared to our earlier proteomics methods, expression of Mx significantly increased with cp BVDV infection only. Mx protein is believed to be induced exclusively via signalling through the type I INF receptor [14,15].

The early stages of the host response to infectious agents include a number of physiological changes, collectively known as the acute phase response. Our previous report identified multiple acute phase response proteins altered by cp BVDV [9]. In this study, acute phase pathway was demonstrated to be the first significant pathway in both ncp and cp BVDV infection. Although, ncp and cp viruses altered different numbers of host proteins in general, they had the same effects on
the monocyte protein expression levels. The acute phase response is comprised of reactions localized at the site of infection, as well as the initiation of s y s t emi c responses, which include a rapid increase in the serum concentration of some proteins, known as acute phase proteins (APP) [16]. Recently, it is becoming clear that viruses interact with iron metabolism. Iron is needed for virus replication, and therefore, by ensuring the infected cell is iron replete, a virus favours its own growth.
Moreover, increased concentrations of iron in the body can cause tissue damage and inflammation and affect organ function [17]. For Hepatitis C viral infection, the
detrimental effects of excess iron are well documented, and elevated iron status is also associated with increased mortality in HIV-1 infection [17]. Here we show that
both cp and ncp BVDV up-regulate transferrin (TF), a negative acute phase protein and a major iron transporter, causing iron overload and exacerbates disease (an animal with an increased serum transferrin level often suffers from iron deficiency anemia). Alternatively, both ncp and cp BVDV down-regulated haptoglobin (HP), a positive acute phase protein capable of binding haemoglobin and removing it from the circulation to prevent iron loss, renal damage and inhibit microbe iron uptake, thus reducing its function as an antioxidant.

Although, High HP levels have been reported in the blood of cattle with infections/diseases like mastitis, metritis, traumatic reticulitis, bacterial nephritis and bovine respiratory syncytial virus [16] and many others,
there is no literature indicating its involvment in BVDV infection. The refore, our finding seems to be unexpected, and to investigate the meaning of these two observations, further studies are needed.
Interstingly, among 69 proteins that have been altered by both biotypes only two proteins, integrin alpha 2b (ITGA2B) and integrin beta 3 (ITGB3), were
differentially altered by cp and ncp BVDV biotypes .

Integrins are the main cell surface receptors for proteins within the extracellular matrix (ECM); they enable cells to migrate, form strong adhesive junctions, and respond to ECM contact by differentiating and/or proliferating [18,19]. Our results indicate that 24h ncp BVDV infect i o n d e c r e a s e d t h e l e v e l s o f I TGA2 B and I TGB 3 , whereas cp BVDV biotype significantly increased their
expression levels. Both integrins are involved in integrin signalling pathway, one of the top ten pathways affected by ncp BVDV-infection. Protein alpha 6 (ITGA6) that
was also down-regulated by ncp BVDV, is known to be a member of the integrin family involved in integrin signalling pathway, and was recently shown to be involved
in cell differentiation [20]. This finding indicates that ncp BVDV unlike the cp counterpart, inhibited the level of communication of the ECM and cell differentiation.
Finally, all the integrins affected by BVDV are also involved in caveolae-mediated endocytosis signalling pathway which was one of the top ten pathways affected
by both ncp and cp BVDV biotypes.
In general, the observed effects of cp BVDV in this study are in agreement with our previois reports suggesting that cp BVDV, while promoting the expression of proteins involved in monocyte activation and differentiation, is inhibiting their antigen presentation to immunocompetent T cells, thus resulting in the uncontrolled inflammation, enhanced viral spread, and impaired anti - viral defense mechanisms in the host.
Unlike the cp BVDV biotype, ncp BVDV increased the expression of proteins involved in compensatory survival and inhibition of cell activation mechanisms, promoting
virus persistence [9,11].

Conclusions

In this study, we identifed bovine proteins whose expression altered significantly during BVDV infection compared to the uninfected monocytes. Those monocyte protein profiles distinguished between the two biotypes showed that cp BVDV had more profound effect
on the protein expression levels with significantly increased number of down - regulated proteins and decreased number of up-regulated proteins compared to
the n cp BVDV biot ype . The use of GO showed profound biotype-related differences in all GO functional groups, indicating that low doses of BVDV infection can
be crucial to understand the complex pathogenesis of BVDV infection. Also, systems biology network modelling identified multiple biotype-related differences in sign if significan biological pathways that could explain the observed biological differences. In particular, our data indicated that only cp BVDV significantly increased the protein expression levels of Mx protein that is believed
to be induced exclusively via signalling through the type I INF receptor. INF receptor signalling activates the adaptive immune responses, and suppression of this signal may be essential for the establishment of persistent
infection that could explain the observed biological differences.

In this study, acute phase pathway was demonstrated to be the first significant pathway in both ncp and cp BVDV infection. Although, ncp and cp viruses altered
different numbers of proteins in general, they had the same effects on the monocyte protein expression levels.
Our finding indicates that n cp BVDV unlike the cp counterpart, inhibited the level of communication of the ECM and cell differentiation thus promoting the establishment of persistent infection. The differences in the expression of the integrins can also mean that cp BVDV infection induces monocytes to differentiate into macrophages or, alternatively that monocytes that have already embarked on the differentiation into macrophages, are more susceptible to cp BVDV infection.
Taken together, the combined use of GO information and systems biology network modelling extended our knowledge of the roles of ncp and cp BVDV biotypes in
the production of persistent infection and cytopathic effects respectively.


Proteomic analysis was carried out with duplicate samples of untreated, cp and ncp-BVDV infected bovine
monocytes using 2D-LC ESI MS
2
as described elsewhere
[23,24]. Briefly, LC analysis was accomplished by strong
cation exchange (SCX) followed by reverse phase (RP)
l iquid chroma tog r aphy ( LC) c oupl ed di r e c t l y in l ine
with electrospray (ESI) ion trap MS. Each DDF fraction
samples from three different infections were loaded into
a LC gradient ion exchange system including a Thermo
Separations P4000 quaternary gradient pump (ThermoEl e c t ron Corpo r a t ion) c oupl ed wi th a 0 .32×100 mm
BioBasic SCX column and run three times. A flow rate
of 3 μl/minwas used for both SCX and RP columns. A
salt gradient was applied in steps of 0, 5, 10, 15, 20, 25,
30, 35, 40, 45, 50, 57, 64, 71, 79, 90, 110, 300, and 700
mM ammonium acetate in 5% acetonitrile, 0.1% formic
acid and the resultant peptides were loaded directly into
the sample loop of a 0.18×100 mm BioBasic C18 RP LC
column of a Proteome X workstation (ThermoElectron).
The RP gradient used 0.1% formic acid in acetonitrile
and increased the acetonitrile concentration in a linear
gradient from 5% to 30% in 30 min and then 30% to
65% in 9 min follow ed by 95% for 5 min and 5% for
15 min.
The spectrum collection time was 59 min for every
SCX step. The LCQ Deca ion trap mass spectrometer
(ThermoElectron) was configured to optimize the duty
cycle length with the quality of data acquired by alternating between a single full MS scan followed by three
tandem MS scans on the three most intense precursor
masses from full scan. The collision energy was normalized to 35%. Dynamic mass exclusion windows were set
at 2 min, and all of the spectra were measured with an
overall mass/charge (m/z) ration range of 200–2000.
Protein identification and differential protein expression
P r o t e i n s w e r e i d e n t i f i e d a nd an a l y z e d a s p r e v i o u s l y [12]. All searches were done using TurboSEQUEST™ (Bioworks Browser 3.2; ThermoElectron) [25].
Mass spectra and tandem mass spectra were searched
against an in silico trypsin-digested non-redundant protein database of Bos taurus downloaded from National
Center for Biotechnology Institute (NCBI). Cysteine carboxyamidomethylation and methionine single and double oxidation were included in the search criteria. Decoy
searches from a randomized version of the bovine protein database were conducted with tandem mass spectra
as described above. The probability for peptide identific a t io n w a s e s t ima t e d u s i n g a me t h o d d e s c r i b e d f o r
Sequest data analysis and was set at p < 0.05 [26]. Probabilities of protein identifications being incorrect were
calculated using published methods [27,28]. Differential
protein expression analysis based on ΣXcorr was carried
out as described in ProtQuant [29]. To correct for mult ipl e t e s t ing , w e d e t e rmined the f a l s e d i s cov e r y r a t e
(FDR) for p value using published methods [30].
Gene Ontology Annotation
Gene ontology (GO) analysis was carried using AgBase
tools [31] to identify the molecular functions, biological
processes and cellular component represented in our
protein datasets. GORetriver tool was used to obtain all
pre-existing GO annotations available for known proteins in our datasets. In addition, we used GOanna to
provide additional GO annotation (i.e. predicted based on s equenc e or tholog e s and ana l y s i s o f func t iona l
domains) for bovine proteins without existing annotation. All GO annotations for our datasets were grouped
into more generalized categories using GOSlimViewer
and summarized using the GOA and Whole Proteome
GOSlim set. Subcategories in each of the three GOSlim
func t i on a l c a t e gor i e s a r e s hown a s a f ol d cha n g e
between the percentages of GO terms identified in cp to

Additional file 1

1. Kobrak A, Weber EL: [Bovine diarrhea virus: an update]. Rev Argent
Microbiol 1997, 29(1):47-61.
2. Houe H: Epidemiological features and economical importance of bovine
virus diarrhoea virus (BVDV) infections. Vet Microbiol 1999, 64(2-3):89-107.
3. Hamers C, Dehan P, Couvreur B, Letellier C, Kerkhofs P, Pastoret PP:
Diversity among bovine pestiviruses. Vet J 2001, 161(2):112-122.
4. Fulton RW, Ridpath JF, Ore S, Confer AW, Saliki JT, Burge LJ, Payton ME:
Bovine viral diarrhoea virus (BVDV) subgenotypes in diagnostic
laboratory accessions: distribution of BVDV1a, 1b, and 2a subgenotypes.
Vet Microbiol 2005, 111(1-2):35-40.
5. Brock KV: The persistence of bovine viral diarrhea virus. Biologicals 2003,
31(2):133-135.
6. Brownlie J, Clarke MC, Howard CJ, Pocock DH: Pathogenesis and
epidemiology of bovine virus diarrhoea virus infection of cattle. Ann
Rech Vet 1987, 18(2):157-166.
7. Brock KV: The many faces of bovine viral diarrhea virus. Vet Clin North
Am Food Anim Pract 2004, 20(1):1-3.
8. Boyd BL, Lee TM, Kruger EF, Pinchuk LM: Cytopathic and non-cytopathic
bovine viral diarrhoea virus biotypes affect fluid phase uptake and
mannose receptor-mediated endocytosis in bovine monocytes. Vet
Immunol Immunopathol 2004, 102(1-2):53-65.
9. Lee SR, Nanduri B, Pharr GT, Stokes JV, Pinchuk LM: Bovine viral diarrhea
virus infection affects the expression of proteins related to professional
antigen presentation in bovine monocytes. Biochim Biophys Acta 2009,
1794(1):14-22.
10. Franchini M, Schweizer M, Matzener P, Magkouras I, Sauter KS, Mirkovitch J,
Peterhans E, Jungi TW: Evidence for dissociation of TLR mRNA
expression and TLR agonist-mediated functions in bovine macrophages.
Vet Immunol Immunopathol 2006, 110(1-2):37-49.
11. Pinchuk GV, Lee SR, Nanduri B, Honsinger KL, Stokes JV, Pinchuk LM: Bovine
viral diarrhea viruses differentially alter the expression of the protein
kinases and related proteins affecting the development of infection and
anti-viral mechanisms in bovine monocytes. Biochim Biophys Acta 2008,
1784(9):1234-1247.
12. Pendarvis K, Kumar R, Burgess SC, Nanduri B: An automated proteomic
data analysis workflow for mass spectrometry. BMC Bioinformatics 2009,
10(Suppl 11):S17.
13. Peterhans E, Jungi TW, Schweizer M: BVDV and innate immunity.
Biologicals 2003, 31(2):107-112
14. von Wussow P, Jakschies D, Hochkeppel HK, Fibich C, Penner L, Deicher H:
The human intracellular Mx-homologous protein is specifically induced
by type I interferons. Eur J Immunol 1990, 20(9):2015-2019.
15. Baigent SJ, Zhang G, Fray MD, Flick-Smith H, Goodbourn S, McCauley JW:
Inhibition of beta interferon transcription by noncytopathogenic bovine
viral diarrhea virus is through an interferon regulatory factor 3-
dependent mechanism. J Virol 2002, 76(18):8979-8988.

16. Heegaard PM, Godson DL, Toussaint MJ, Tjornehoj K, Larsen LE, Viuff B,
Ronsholt L: The acute phase response of haptoglobin and serum
amyloid A (SAA) in cattle undergoing experimental infection with
bovine respiratory syncytial virus. Vet Immunol Immunopathol 2000, 77(1-
2):151-159.
17. Drakesmith H, Prentice A: Viral infection and iron metabolism. Nat Rev
Microbiol 2008, 6(7):541-552.
18. Delon I, Brown NH: Integrins and the actin cytoskeleton. Curr Opin Cell
Biol 2007, 19(1):43-50.
19. Le Clainche C, Carlier MF: Regulation of actin assembly associated with
protrusion and adhesion in cell migration. Physiol Rev 2008,
88(2):489-513.
20. Meighan CM, Schwarzbauer JE: Temporal and spatial regulation of
integrins during development. Curr Opin Cell Biol 2008, 20(5):520-524.
21. Lee SR, Pharr GT, Boyd BL, Pinchuk LM: Bovine viral diarrhea viruses
modulate toll-like receptors, cytokines and co-stimulatory molecules
genes expression in bovine peripheral blood monocytes. Comp Immunol
Microbiol Infect Dis 2008, 31(5):403-418.
22. Kruger EF, Boyd BL, Pinchuk LM: Bovine monocytes induce
immunoglobulin production in peripheral blood B lymphocytes. Dev
Comp Immunol 2003, 27(10):889-897.
23. Lee SR, Pharr GT, Cooksey AM, McCarthy FM, Boyd BL, Pinchuk LM:
Differential detergent fractionation for non-electrophoretic bovine
peripheral blood monocyte proteomics reveals proteins involved in
professional antigen presentation. Dev Comp Immunol 2006,
30(11):1070-1083.
24. McCarthy FM, Burgess SC, van den Berg BH, Koter MD, Pharr GT:
Differential detergent fractionation for non-electrophoretic eukaryote
cell proteomics. J Proteome Res 2005, 4(2):316-324.
25. Eng JK, McCormack AL, Yates JR: An approach to correlate tandem mass
spectral data of peptides with amino acid sequences in a protein
database. J Am Soc Mass Spectrom 1994, 5:976-989.
26. Qian WJ, Liu T, Monroe ME, Strittmatter EF, Jacobs JM, Kangas LJ, Petritis K,
Camp DG, Smith RD 2nd: Probability-based evaluation of peptide and
protein identifications from tandem mass spectrometry and SEQUEST
analysis: the human proteome. J Proteome Res 2005, 4(1):53-62.
27. Lopez-Ferrer D, Martinez-Bartolome S, Villar M, Campillos M, MartinMaroto F, Vazquez J: Statistical model for large-scale peptide
identification in databases from tandem mass spectra using SEQUEST.
Anal Chem 2004, 76(23):6853-6860.
28. MacCoss MJ, Wu CC, Yates JR 3rd: Probability-based validation of protein
identifications using a modified SEQUEST algorithm. Anal Chem 2002,
74(21):5593-5599.
29. Bridges SM, Magee GB, Wang N, Williams WP, Burgess SC, Nanduri B:
ProtQuant: a tool for the label-free quantification of MudPIT proteomics
data. BMC Bioinformatics 2007, 8(Suppl 7):S24.
30. Benjamini Y, Hochberg Y: Controlling the false discovery rate: a practical
and powerful approach to multiple testing. Journal of the Royal Statistical
Society: Series B (Statistical Methodology) 1995, 57:289-300.
31. McCarthy FM, Bridges SM, Wang N, Magee GB, Williams WP, Luthe DS,
Burgess SC: AgBase: a unified resource for functional analysis in
agriculture. Nucleic Acids Res 2007, 35(Database issue):D599-603.
32. Gerling IC, Singh S, Lenchik NI, Marshall DR, Wu J: New data analysis and
mining approaches identify unique proteome and transcriptome
markers of susceptibility to autoimmune diabetes. Mol Cell Proteomics
2006, 5(2):293-305.
33. Peddinti D, Nanduri B, Kaya A, Feugang JM, Burgess SC, Memili E:
Comprehensive proteomic analysis of bovine spermatozoa of varying
fertility rates and identification of biomarkers associated with fertility.
BMC Syst Biol 2008, 2:19.