Volume 15 Supplement 1
An improved ChIP-seq peak detection system for simultaneously identifying post-translational modified transcription factors by combinatorial fusion, using SUMOylation as an example
© Cheng et al.; licensee BioMed Central Ltd. 2014
Published: 24 January 2014
Post-translational modification (PTM) of transcriptional factors and chromatin remodelling proteins is recognized as a major mechanism by which transcriptional regulation occurs. Chromatin immunoprecipitation (ChIP) in combination with high-throughput sequencing (ChIP-seq) is being applied as a gold standard when studying the genome-wide binding sites of transcription factor (TFs). This has greatly improved our understanding of protein-DNA interactions on a genomic-wide scale. However, current ChIP-seq peak calling tools are not sufficiently sensitive and are unable to simultaneously identify post-translational modified TFs based on ChIP-seq analysis; this is largely due to the wide-spread presence of multiple modified TFs. Using SUMO-1 modification as an example; we describe here an improved approach that allows the simultaneous identification of the particular genomic binding regions of all TFs with SUMO-1 modification.
Traditional peak calling methods are inadequate when identifying multiple TF binding sites that involve long genomic regions and therefore we designed a ChIP-seq processing pipeline for the detection of peaks via a combinatorial fusion method. Then, we annotate the peaks with known transcription factor binding sites (TFBS) using the Transfac Matrix Database (v7.0), which predicts potential SUMOylated TFs. Next, the peak calling result was further analyzed based on the promoter proximity, TFBS annotation, a literature review, and was validated by ChIP-real-time quantitative PCR (qPCR) and ChIP-reChIP real-time qPCR. The results show clearly that SUMOylated TFs are able to be pinpointed using our pipeline.
A methodology is presented that analyzes SUMO-1 ChIP-seq patterns and predicts related TFs. Our analysis uses three peak calling tools. The fusion of these different tools increases the precision of the peak calling results. TFBS annotation method is able to predict potential SUMOylated TFs. Here, we offer a new approach that enhances ChIP-seq data analysis and allows the identification of multiple SUMOylated TF binding sites simultaneously, which can then be utilized for other functional PTM binding site prediction in future.
SUMOylation was initially identified as a reversible post-translational modification that controls a variety of cellular processes including replication, chromosome segregation, and DNA repair [1–3]. The growing list of SUMO substrates includes various transcription factors (TFs) and chromatin remodeling molecules, which, upon SUMOylation, are often associated with transcriptional repression , and the maintenance of heterochromatin silencing [5, 6]. The deregulation of SUMOylation has been associated with a number of diseases including cancer [7–10]. SUMO has been found in all eukaryotes, but not in prokaryotes. Furthermore, the global regulatory role of SUMO in gene expression and protein interactions has been shown to be richly exploited in lower eukaryotes such as yeast [11, 12]. While numerous studies have provided considerable insight into the regulation of SUMOylated proteins in higher eukaryotes, their scope usually has been limited to a single host factor. The underlying complexity of SUMOylation has been extended by identifying the downstream consequences of these non-covalent interactions with effectors via SUMO interaction motifs (SIMs) , with the SIMs being critical to both SUMO conjugation and SUMO-mediated effects. Exploring the functions of SUMO conjugation and interaction during epigenetic regulation in mammalian cells will considerably enhance our knowledge of transcriptional regulation of SUMOylation in higher eukaryotes.
SUMOylation of transcriptional regulators results in alterations to the transcription regulation of individual genes, while the SUMOylation of epigenetic regulators brings about long-range chromatin remodeling, and hence global changes in expression. When chromatin structures are regulated by SUMO, it has been found to involve SUMO targeting of histone deacetylases and this then results in histone deacetylation, chromosome condensation, and transcriptional repression. At the same time, numerous transcription factors have been reported to be SUMO substrates, including Elk-1, SP1 , AP2, and many others. The study of epigenetic regulation when there is PTM of regulatory transcription factors is still in its infancy and there remains a need for new and improved screening tools as well as the development of assay pipelines.
Recently, chromatin immunoprecipitation (ChIP) followed by high-throughput sequencing (ChIP-seq), has become a powerful and high resolution method that allows the study of the impact of TFs and their co-regulators in higher eukaryotes in a genome-wide manner [17, 18]. During the ChIP process, DNA is initially cross-linked in a specific sample to the protein that binds to it. This cross-linked DNA is then broken into fragments and immunoprecipitation with a specific antibody for the DNA-binding protein follows; finally, the associated DNA is identified after de-crosslinking. High-throughput sequencing of short tags (reads) can be achieved using the resulting DNA population. ChIP-seq involves the short read (30~100 bp) sequencing of ChIP-enriched DNA fragments. These reads are subsequently aligned to a reference genome such as hg19. The first step of all ChIP-seq analyses is peak detection. Peaks are regions that are markedly enriched in terms of read density based on the ChIP-seq data. Potential transcription factor binding sites (TFBS) can only be precisely identified when the true peaks are detected first by "peak calling" tools.
Global identification of SUMO-1 peaks in a primary effusion lymphoma (PEL) cell line, BCBL-1
SUMO-1 ChIP-seq alignment results
SUMO-1 ChIP-seq data
# of total reads
# of filtered reads
# of duplicate reads
# of mapped reads
Peak features obtained using the individual, union and intersection methods
TFBS peak #
MACS = M
CisGenome = C
T-PIC = T*
BayesPeak = B
Intersection of different peak calling tools represents positive results
Precision indices for the single methods
Precision indices for the union (+) of two or three methods
Precision indices for the intersection (*) of two or three methods
Validation the data from ChIP-seq for ELK-1 binding sites with SUMO-1 enrichment by real-time qPCR
Potential SUMO-1 targeting TF identification that relies on SUMO-1ChIP peak height scores and can be validated via a literature review
Potential SUMO-1 TF list
Name of TF
Among the 19 potential SUMO TFs, 17 of them have been previously identified as SUMO substrates. For example, Elk-1, the top 1 SUMOylated TF candidate in our analysis, can be SUMO modified at its R motif . Overall, 30% of the SUMO peaks (149/482) containing the Elk-1 TFBS that were identified in the present study are also found in another Elk-1 ChIP-seq data (GEO ACCESSION: GSM608163). Although no previous study reports have indicated that hRFX1 and NSCL1 are SUMOylated, we cannot rule out the potential of these two proteins to form a SUMO complex and/or to bind a SUMOylated cofactor.
Validation of SUMO-1 enrichment in ELK-1 binding site identified in HeLa cells
Recently, a ChIP-seq report has pinpointed the global chromatin localization of ELK-1 in human HeLa cells (GEO ACCESSION: GSM608163). We selected three high quality ELK-1 binding sites identified in HeLa cells overlapping with our SUMO-1 enriched regions and validated by ChIP-qPCR. As shown in Figure 3D to 3F, there is a significant increase of SUMO-1 enrichment in ELK-1 binding sites of SNRPE, INO80B and LYSMD1 promoter identified from previous study of others in HeLa cells. However, ChIP-reChIP data shows no co-localization of ELK-1 and SUMO-1 in the promoter region of SNRPE, INO80B and LYSMD1 genes (Figure 4C to 4E). Consistent with the ChIP-reChIP data, the transcription of all three genes showed no changes during K-Rta induced KSHV reactivation after SUMO-1 knockdown comparing to the control cells (Figure 5D to 5F). The results were similar to the negative control genes, MCL-1 and IRF-3, which have ELK-1 binding site but not SUMO-1 enrichment in their promoter regions (Figure 5G and 5H). The inconsistency between our results and the findings in HeLa cells may be due to the cell type specificity. Together, taking ELK-1 as an example, the result here suggests that our pipeline is able to identify the potential chromatin region bound by SUMO modified transcription factors successfully. The biological role of SUMOylation in regulating the function of ELK-1 was also identified in a cell type-specific manner.
Comparisons of the different methods available for the global identification of SUMO-1 peaks
Evaluation of system fusion result
Decoding how PTMs impacts the TF regulatory system that governs diverse cellular responses remains challenging. Taking SUMO modification as an example, we have developed a computational pipeline for predicting putative SUMOylated TFs from a group of TFBS. Using the combinatorial fusion methods described here, there is no need to depend on a single "best" algorithm. The merge detection method is able to find peaks with greater accuracy than any other peak calling software alone using ChIP-seq data retrieved from targeted PTMs. SUMO-1 target TFs are predicted well using our pipeline. In total, 89% of the 19 potential SUMOylated TFs were found to be SUMOylated after confirmation by literature review. In summary, our observations includes: (1) based on the criteria and performance evaluations used, there are no single answer to the selection of the best available method for ChIP-seq peak detection when identifying PTMs; (2) combinations of different tools are able to improve results in many cases; and (3) M*C*T is the superior prediction method when detecting putative SUMOylated TFs. More than 60% of the peaks identified in this study have not been annotated. One of the reasons for this is that the human cell contains thousands of TFs, and many of them are able to be SUMOylated. The TFBS data set from the UCSC genome browser only includes binding sites for 258 TFs out of these thousands of TFs. In the future, our work should help researchers to achieve a greater understanding of SUMOylated TFs once a better TFBS database become available. Moreover, we intend to explore the non-TFBS-annotated SUMO peaks in order to identify chromatin remodeling molecules that are not TFs. Most importantly, our pipeline here provides a platform for all researchers who want to study the relation between PTM and epigenetic regulation using their own chromatin binding data.
Materials and methods
Experiment design and sample preparation
The epigenetic study underling this paper's aim is an investigation of the impact of SUMO/TF interaction in a primary effusion lymphoma (PEL) cell line, BCBL-1. To this end, we generated ChIP-seq data using anti-SUMO-1 antibodies and BCBL-1. In general, the results of a SUMO-1 ChIP-seq experiment were anticipated to reflect indirectly the SUMO regulatory genome via SUMOylated TF binding and chromatin. In parallel to this, another scenario is that SUMO-1 antibody identifies SUMOylated cofactors that are recruited to TFs and TF-occupied DNA sequences. The cross-linked SUMO-TF-DNA complexes were extracted and contained the DNA that interacts with either the SUMOylated-TFs or the SUMOylated transcription complexes. After purification of ChIP-enriched DNA, a library was developed to allow sequencing on a NGS platform (Figure 1).
KSHV infected primary effusion lymphoma (PEL) cell line, TREx-F3H3-K-Rta BCBL-1 was grown in RPMI 1640 containing 15% FBS, 50 μg/ml blasticidin and 100 μg/ml hygromycin(Invitrogen, Carlsbad, CA) in the presence of 5% CO2.
The shRNA cassette of SUMO-1 (5'-CACCCAACACATCTCAAGAAACTCACGAATGAGTTTCTTGAGATGTGTTG-3') was inserted into pLenti4-T/O-shRNA plasmid and introduced into TREx-F3H3-K-Rta BCBL-1 cells by lentiviral transduction. Cells were selected for 14 days by 300 μg/ml zeocine (InvivoGen, ant-zn-1) and purified by Ficoll. Knockdown efficiency of SUMO-1 by shRNA were tested by treated the cells with doxycycline (Dox) for 48 hours. TREx-F3H3-K-Rta-shSUMO-1 BCBL-1 cells were maintained as described for TREx-F3H3-K-Rta BCBL-1 and supplemented with 300 μg/ml zeocine.
Chromatin immunoprecipitation-sequencing (ChIP-Seq), ChIP-reChIP assay and real-time quantitative PCR (qPCR)
1 × 107 cells were harvested and ChIP assays were performed following the protocol described by the Farnham laboratory (provided at http://genomics.ucdavis.edu/farnham). ChIP-reChIP assays were performed by Re-ChIP-IT kit (Active Motif, Carlsbad, CA) following the manufacturer's instruction. ChIP-verified rabbit polyclone antibodies specific against SUMO-1 (Abcam, Cambridge, MA) and rabbit non-immune serum IgG (Alpha Diagnostic International) were used for the ChIP and ChIP-reChIP assays.
ChIP-seq library construction was carried out following the sample preparation protocol from Illumina. Short reads (100 bp) from both ends (paired-end sequencing) were sequenced on an Illumina® Genome AnalyzerIIX System. The binding sites were verified by SYBR® Green Based qPCR using a CFX connect™ real-time PCR detection system (Bio-Rad, Richmond, CA). Specific primer sets were designed around the identified binding sites for this purpose. Enrichment of SUMO-1 and IgG samples were normalized with the input.
The reads within the SUMO-1 ChIP-seq data sets were aligned by BWA with default parameters . hg19 was used as the reference genome, having been downloaded from the UCSC genome browser . The Ensembl database was used to define gene regions . Promoter regions are defined as the area that stretches from 5 kb upstream to 1kb downstream of a transcription start site (TSS).
Scoring system for TFBS in SUMO peaks
It means the sum of the peak heights D i x weighting with the percentage of D i x in all TB i . Let the rank function R(D i x ) be the function from 1 to 258 that is obtained by sorting the values in S(D i x ) into descending order and converting the function S(TB x ) into the function R(TB x ) using the rank as its function value.
Combined two peak detection systems
and R(D x+y ) is the rank function obtained as R(D x+y ).
The intersection of two system, X and Y, D x*y is the set of SUMO TFBS that are detected both by X and Y.
D x*y ⊆ D x+y where D x+y = D x ∩D y
and R(D x*y ) is the rank function obtained as R(D x*y ).
Identifying potential SUMO-1 target TFs using the Hampel Identifier
where, M is the median of , ,... observations.
MADN = MAD/0.6745, and MAD is the median of the values , ,..., . 0.6745 is 0.75 quantile of standard normal distribution, and 2.24 is 0.975 quantile of chi-square distribution with one degree of freedom.
Performance evaluation methods
Conflicts of interest
The authors declare that they have no competing interests.
This study was supported by NSC grants (101-2321-B-010-017 and 101-2325-B-007-001) and by NHRI grants (NHRI-EX102-10215BC). We thank Biomedical Science and Engineering Center at National Tsing Hua University for the ChIP-Seq experiments.
Publication of this article was funded by NSC grants (101-2321-B-010-017).
This article has been published as part of BMC Genomics Volume 15 Supplement 1, 2014: Selected articles from the Twelfth Asia Pacific Bioinformatics Conference (APBC 2014): Genomics. The full contents of the supplement are available online at http://www.biomedcentral.com/bmcgenomics/supplements/15/S1.
- Lomeli H, Vazquez M: Emerging roles of the SUMO pathway in development. Cellular and molecular life sciences: CMLS. 2011, 68 (24): 4045-4064. 10.1007/s00018-011-0792-5.PubMedView ArticleGoogle Scholar
- Nacerddine K, Lehembre F, Bhaumik M, Artus J, Cohen-Tannoudji M, Babinet C, Pandolfi PP, Dejean A: The SUMO pathway is essential for nuclear integrity and chromosome segregation in mice. Developmental cell. 2005, 9 (6): 769-779. 10.1016/j.devcel.2005.10.007.PubMedView ArticleGoogle Scholar
- Prudden J, Perry JJ, Nie M, Vashisht AA, Arvai AS, Hitomi C, Guenther G, Wohlschlegel JA, Tainer JA, Boddy MN: DNA repair and global sumoylation are regulated by distinct Ubc9 noncovalent complexes. Molecular and cellular biology. 2011, 31 (11): 2299-2310. 10.1128/MCB.05188-11.PubMedPubMed CentralView ArticleGoogle Scholar
- Garcia-Dominguez M, Reyes JC: SUMO association with repressor complexes, emerging routes for transcriptional control. Biochimica et biophysica acta. 2009, 1789 (68): 451-459.PubMedView ArticleGoogle Scholar
- Wilkinson KA, Henley JM: Mechanisms, regulation and consequences of protein SUMOylation. The Biochemical journal. 2010, 428 (2): 133-145. 10.1042/BJ20100158.PubMedPubMed CentralView ArticleGoogle Scholar
- Maison C, Bailly D, Roche D, Montes de Oca R, Probst AV, Vassias I, Dingli F, Lombard B, Loew D, Quivy JP, et al: SUMOylation promotes de novo targeting of HP1alpha to pericentric heterochromatin. Nature genetics. 2011, 43 (3): 220-227. 10.1038/ng.765.PubMedView ArticleGoogle Scholar
- Driscoll JJ, Pelluru D, Lefkimmiatis K, Fulciniti M, Prabhala RH, Greipp PR, Barlogie B, Tai YT, Anderson KC, Shaughnessy JD, et al: The sumoylation pathway is dysregulated in multiple myeloma and is associated with adverse patient outcome. Blood. 2010, 115 (14): 2827-2834. 10.1182/blood-2009-03-211045.PubMedPubMed CentralView ArticleGoogle Scholar
- Kim JH, Lee JM, Nam HJ, Choi HJ, Yang JW, Lee JS, Kim MH, Kim SI, Chung CH, Kim KI, et al: SUMOylation of pontin chromatin-remodeling complex reveals a signal integration code in prostate cancer cells. Proceedings of the National Academy of Sciences of the United States of America. 2007, 104 (52): 20793-20798. 10.1073/pnas.0710343105.PubMedPubMed CentralView ArticleGoogle Scholar
- Baek SH: A novel link between SUMO modification and cancer metastasis. Cell Cycle. 2006, 5 (14): 1492-1495. 10.4161/cc.5.14.3008.PubMedView ArticleGoogle Scholar
- Bawa-Khalfe T, Yeh ET: SUMO Losing Balance: SUMO Proteases Disrupt SUMO Homeostasis to Facilitate Cancer Development and Progression. Genes & cancer. 2010, 1 (7): 748-752. 10.1177/1947601910382555.View ArticleGoogle Scholar
- Rosonina E, Duncan SM, Manley JL: SUMO functions in constitutive transcription and during activation of inducible genes in yeast. Genes & development. 2010, 24 (12): 1242-1252. 10.1101/gad.1917910.View ArticleGoogle Scholar
- Makhnevych T, Sydorskyy Y, Xin X, Srikumar T, Vizeacoumar FJ, Jeram SM, Li Z, Bahr S, Andrews BJ, Boone C, et al: Global map of SUMO function revealed by protein-protein interaction and genetic networks. Molecular cell. 2009, 33 (1): 124-135. 10.1016/j.molcel.2008.12.025.PubMedView ArticleGoogle Scholar
- Lin DY, Huang YS, Jeng JC, Kuo HY, Chang CC, Chao TT, Ho CC, Chen YC, Lin TP, Fang HI, et al: Role of SUMO-interacting motif in Daxx SUMO modification, subnuclear localization, and repression of sumoylated transcription factors. Molecular cell. 2006, 24 (3): 341-354. 10.1016/j.molcel.2006.10.019.PubMedView ArticleGoogle Scholar
- Yang SH, Sharrocks AD: PIASxalpha differentially regulates the amplitudes of transcriptional responses following activation of the ERK and p38 MAPK pathways. Molecular cell. 2006, 22 (4): 477-487. 10.1016/j.molcel.2006.03.037.PubMedView ArticleGoogle Scholar
- Spengler ML, Brattain MG: Sumoylation inhibits cleavage of Sp1 N-terminal negative regulatory domain and inhibits Sp1-dependent transcription. The Journal of biological chemistry. 2006, 281 (9): 5567-5574.PubMedView ArticleGoogle Scholar
- Verger A, Perdomo J, Crossley M: Modification with SUMO. A role in transcriptional regulation. EMBO reports. 2003, 4 (2): 137-142. 10.1038/sj.embor.embor738.PubMedPubMed CentralView ArticleGoogle Scholar
- Kharchenko PV, Tolstorukov MY, Park PJ: Design and analysis of ChIP-seq experiments for DNA-binding proteins. Nature biotechnology. 2008, 26 (12): 1351-1359. 10.1038/nbt.1508.PubMedPubMed CentralView ArticleGoogle Scholar
- Whitington T, Frith MC, Johnson J, Bailey TL: Inferring transcription factor complexes from ChIP-seq data. Nucleic acids research. 2011, 39 (15): e98-10.1093/nar/gkr341.PubMedPubMed CentralView ArticleGoogle Scholar
- He Y, Zhang Y, Zheng G, Wei C: CTF: a CRF-based transcription factor binding sites finding system. BMC Genomics. 2012, 13 (Suppl 8): S18-10.1186/1471-2164-13-S8-S18.PubMedPubMed CentralView ArticleGoogle Scholar
- Pepke S, Wold B, Mortazavi A: Computation for ChIP-seq and RNA-seq studies. Nature methods. 2009, 6 (11 Suppl): S22-32.PubMedPubMed CentralView ArticleGoogle Scholar
- Laajala TD, Raghav S, Tuomela S, Lahesmaa R, Aittokallio T, Elo LL: A practical comparison of methods for detecting transcription factor binding sites in ChIP-seq experiments. BMC genomics. 2009, 10:Google Scholar
- Wilbanks EG, Facciotti MT: Evaluation of algorithm performance in ChIP-seq peak detection. PloS one. 2010, 5 (7): e11471-10.1371/journal.pone.0011471.PubMedPubMed CentralView ArticleGoogle Scholar
- Oh YM, Kim JK, Choi S, Yoo JY: Identification of co-occurring transcription factor binding sites from DNA sequence using clustered position weight matrices. Nucleic acids research. 2012, 40 (5): e38-10.1093/nar/gkr1252.PubMedPubMed CentralView ArticleGoogle Scholar
- Bardet AF, He Q, Zeitlinger J, Stark A: A computational pipeline for comparative ChIP-seq analyses. Nature protocols. 2012, 7 (1): 45-61.View ArticleGoogle Scholar
- Zhang Y, Liu T, Meyer CA, Eeckhoute J, Johnson DS, Bernstein BE, Nusbaum C, Myers RM, Brown M, Li W, et al: Model-based analysis of ChIP-Seq (MACS). Genome biology. 2008, 9 (9): R137-10.1186/gb-2008-9-9-r137.PubMedPubMed CentralView ArticleGoogle Scholar
- Hower V, Evans SN, Pachter L: Shape-based peak identification for ChIP-Seq. BMC bioinformatics. 2011, 12: 15-10.1186/1471-2105-12-15.PubMedPubMed CentralView ArticleGoogle Scholar
- Cairns J, Spyrou C, Stark R, Smith ML, Lynch AG, Tavare S: BayesPeak--an R package for analysing ChIP-seq data. Bioinformatics. 2011, 27 (5): 713-714. 10.1093/bioinformatics/btq685.PubMedPubMed CentralView ArticleGoogle Scholar
- Ji H, Jiang H, Ma W, Wong WH: Using CisGenome to analyze ChIP-chip and ChIP-seq data. Current protocols in bioinformatics/editoral board, Andreas D Baxevanis [et al]. 2011, Chapter 2:Unit2 13Google Scholar
- Schweikert C, Brown S, Tang Z, Smith PR, Hsu DF: Combining multiple ChIP-seq peak detection systems using combinatorial fusion. BMC genomics. 2012, S12-13 Suppl 8
- Yang SH, Jaffray E, Hay RT, Sharrocks AD: Dynamic interplay of the SUMO and ERK pathways in regulating Elk-1 transcriptional activity. Molecular cell. 2003, 12 (1): 63-74. 10.1016/S1097-2765(03)00265-X.PubMedView ArticleGoogle Scholar
- Yang L, Lin C, Liu W, Zhang J, Ohgi KA, Grinstein JD, Dorrestein PC, Rosenfeld MG: ncRNA- and Pc2 methylation-dependent gene relocation between nuclear structures mediates gene activation programs. Cell. 2011, 147 (4): 773-788. 10.1016/j.cell.2011.08.054.PubMedPubMed CentralView ArticleGoogle Scholar
- Arito M, Horiba T, Hachimura S, Inoue J, Sato R: Growth factor-induced phosphorylation of sterol regulatory element-binding proteins inhibits sumoylation, thereby stimulating the expression of their target genes, low density lipoprotein uptake, and lipid synthesis. The Journal of biological chemistry. 2008, 283 (22): 15224-15231. 10.1074/jbc.M800910200.PubMedPubMed CentralView ArticleGoogle Scholar
- Ryan CM, Kindle KB, Collins HM, Heery DM: SUMOylation regulates the nuclear mobility of CREB binding protein and its association with nuclear bodies in live cells. Biochemical and biophysical research communications. 2010, 391 (1): 1136-1141. 10.1016/j.bbrc.2009.12.040.PubMedPubMed CentralView ArticleGoogle Scholar
- Flicek P, Amode MR, Barrell D, Beal K, Brent S, Carvalho-Silva D, Clapham P, Coates G, Fairley S, Fitzgerald S, et al: Ensembl 2012. Nucleic acids research. 2012, 40 (Database): D84-90.PubMedPubMed CentralView ArticleGoogle Scholar
- Meyer LR, Zweig AS, Hinrichs AS, Karolchik D, Kuhn RM, Wong M, Sloan CA, Rosenbloom KR, Roe G, Rhead B, et al: The UCSC Genome Browser database: extensions and updates 2013. Nucleic acids research. 2013, 41 (Database): D64-69.PubMedPubMed CentralView ArticleGoogle Scholar
- Davies L, Gather U: The identification of multiple outliers. Journal of the American Statistical Association. 1993, 88
- Perarson RK: Outliers in process modeling and identification. IEEE Transactions On Control Systems Technology. 2002, 10:Google Scholar
- Davies L, Gather U: The Identification of Multiple Outliers. Journal of the American Statistical Association. 1993, 88 (423): 782-792. 10.1080/01621459.1993.10476339.View ArticleGoogle Scholar
- Li H, Durbin R: Fast and accurate long-read alignment with Burrows-Wheeler transform. Bioinformatics. 2010, 26 (5): 589-595. 10.1093/bioinformatics/btp698.PubMedPubMed CentralView ArticleGoogle Scholar
- Ji Z, Degerny C, Vintonenko N, Deheuninck J, Foveau B, Leroy C, Coll J, Tulasne D, Baert JL, Fafeur V: Regulation of the Ets-1 transcription factor by sumoylation and ubiquitinylation. Oncogene. 2007, 26 (3): 395-406. 10.1038/sj.onc.1209789.PubMedView ArticleGoogle Scholar
- Begitt A, Droescher M, Knobeloch KP, Vinkemeier U: SUMO conjugation of STAT1 protects cells from hyperresponsiveness to IFNgamma. Blood. 2011, 118 (4): 1002-1007. 10.1182/blood-2011-04-347930.PubMedView ArticleGoogle Scholar
- Hamard PJ, Boyer-Guittaut M, Camuzeaux B, Dujardin D, Hauss C, Oelgeschlager T, Vigneron M, Kedinger C, Chatton B: Sumoylation delays the ATF7 transcription factor subcellular localization and inhibits its transcriptional activity. Nucleic acids research. 2007, 35 (4): 1134-1144. 10.1093/nar/gkl1168.PubMedPubMed CentralView ArticleGoogle Scholar
- Lu D, Han C, Wu T: Microsomal prostaglandin E synthase-1 promotes hepatocarcinogenesis through activation of a novel EGR1/beta-catenin signaling axis. Oncogene. 2012, 31 (7): 842-857. 10.1038/onc.2011.287.PubMedPubMed CentralView ArticleGoogle Scholar
- Iwasaki K, Hailemariam K, Tsuji Y: PIAS3 interacts with ATF1 and regulates the human ferritin H gene through an antioxidant-responsive element. The Journal of biological chemistry. 2007, 282 (31): 22335-22343. 10.1074/jbc.M701477200.PubMedPubMed CentralView ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.