Results: Principal component analysis using 84 genes showed that the patient sample was clearly clustering with neuroblastoma tumors. This was confirmed by hierarchical clustering of the multiplex RT-PCR data. The patient under- went treatment for high-risk neuroblastoma comprising chemotherapy including cisplatin, etoposide, vindesine, dacarbacine, ifosfamide, vincristine, adriamycine and autologous stem cell transplantation followed by maintenance therapy with 13-cis retinoic acid (GPOH NB2004 High Risk Trial Protocol) and is in complete long-term remission. Conclusion: The use of gene expressionprofiling in an individual patient strongly contributed to clarification in a diagnostic dilemma which finally led to a change of diagnosis from nephroblastoma to neuroblastoma. This case underlines the importance of gene-expressionprofiling in the correct diagnosis of childhood neoplasms with atypical presentation to ensure that adequate treatment regimens can be applied.
Members of the Klotho gene family have been identified as modulators of the aging process. Deletion of αklotho in the mouse results in a syndrome resembling rapid human aging. Conversely, overexpression of αklotho extends mammalian lifespan. Here, we identify klotho orthologs in the vertebrate aging model Nothobranchius furzeri and provide a detailed spatio-temporal expression profile of both paralogs, α and βklotho, from embryogenesis until old age spanning the entire life cycle of the organism. Specifically, we observe low levels of expression of both paralogs during embryogenesis followed by a significant transcriptional induction as development proceeds. In adult killifish, αklotho is predominantly expressed in the liver, the kidney, and the developing pharyngeal teeth. Particularly high levels of αKlotho protein were identified in the kidney tubules, closely resembling mammalian expression patterns. Prominent βklotho expression was detected in the killifish intestine and liver. Overall, qRT-PCR analysis of Klotho members as a function of age revealed steady transcript levels, except for βklotho expression in the liver which was significantly downregulated with age. This spatio-temporal expressionprofiling may serve as a useful starting point to further investigate the distinct physiological roles of Klotho members during the aging process. Keywords Klotho . African killifish . Nothobranchius furzeri . Aging
related to pancytopenia due to bone marrow involvement as well as weight loss and pain more frequently. Never- theless, in some cases it may be difficult to distinguish neuroblastoma from nephroblastoma with standard diagnostic procedures but this is crucial with regard to optimal therapeutical management of these tumors [ 2 , 3 ]. We describe a 4½ year old girl with neuroblastoma mis- diagnosed as nephroblastoma based on the constellation of radiological and laboratory findings. Gene expressionprofiling and multiplex RT-PCR strongly supported the final diagnosis of neuroblastoma, underlining the impor- tance of including molecular techniques in the diagnosis of childhood neoplasms with atypical presentation. Case presentation
Expressionprofiling was performed on Affymetrix MG U74 Chips. Analyzing the data, the hierarchical clustering resulted in a conclusion that there was a high degree of similarity between IV group and control group for both the models suggesting that PLY application via IV mode does not alter gene expression pattern considerably. The reason for this finding might be that PLY in the given dosage is completely bound by serum proteins and therefore does not develop effects. Consequently, we put IV and control in one group. Afterwards, differential expression of IT group versus control+IV group was calculated. As expected, we found several genes that are regulating in PLY dependent manner. Lysyl oxidase (LOX), an enzyme secreted by activated vascular smooth muscle cells and fibroblasts, was upregulated with a more than a factor of 4.0. LOX catalyzes a key step in the cross-linking and stabilization of collagen and elastin in the vascular wall and involved in extracellular matrix maturation. Upregulation of LOX could be a sign for instability between collagen and elastin in vascular wall and impairs the endothelial barrier function and could be involved in homocysteine (HC)-induced endothelial dysfunction and leakage. 124
Several things should be considered before using anti-mir and pre-mir, the first is to protect synthetic anti-miRs and pre-miRs from cellular nucleases when applied to cell culture, the second is to improve affinity for the target miRNA (in case of anti-miRs) or mRNA (in case of pre-miRs) and enhanced RNase H or other enzyme activity that is involved in miRNA function (Esau 2008). To enhance novel miRNAs, instead of using mimic miRNA, we cloned pre-miRNAs into a destination vector. In order to improve affinity for the target miRNAs, we applied LNA probes. The use of LNA probes for the knockdown of miRNA is particularly effective because these probes have high affinity for their short RNA targets and allow the discrimination of closely related miRNA sequences. Next, we performed qRT-PCR to evaluate functionality of pre-miRs and anti-miRs in cell culture. MiRNA expressionprofiling by qRT-PCR shows that transfection with pre-miRs significantly increased the expression of both novel miRNAs in SU- DHL-4 and Raji cell lines. Except for NB-miR-19 in SU-DHL-4 cell line, we observed a significant reduction of both novel miRNAs levels in cells transfected with LNA probes. It is worth nothing that, sometimes inhibition of miRNAs is achieved without detectable miRNA degradation (Elmen, Lindow et al. 2008, Davis, Propp et al. 2009), suggesting that qRT-PCR might not always be an appropriate method to measure inhibition. Furthermore, in some cases, high affinity chemical modifications make the anti-miR:miRNA complex stable and interfere with miRNA detection by qRT-PCR (Davis, Propp et al. 2009). However, compared to prior studies (Davis, Lollo et al. 2006, Ovcharenko, Kelnar et al. 2007, Lennox and Behlke 2010), we confirmed that synthetic pre-miR and anti-miR are functional methods to, respectively, increase or inhibit the activity of endogenous miRNAs in cultured cells.
Vor über 10 Jahren wurde das Fachgebiet Proteomics definiert. Viele der verwendeten Technologien wurden seither weiter bzw. neu entwickelt und in zahlreichen Studien zur Klärung biologischer Fragestellungen verwendet. In den Proteom-Analysen werden große Datenmengen erzeugt. Die Stringenz, die für die Erhebung der Daten erforderlich ist, wurde in einigen Studien suboptimal berücksichtigt, so dass die Aussagekraft einiger Publikationsdaten rückwirkend überprüft und/oder validiert werden müssen (Wilkins et al., 2006). Um die Qualität der generierten Informationen zu erhöhen, müssen in zukünftigen Studien die Art und Anzahl der verwendeten Replikate, die Reproduzierbarkeit der Daten sowie das experimentelle Design besser berücksichtigt werden (Wilkins et al., 2006; Karp & Lilley, 2007). Eine Bewertung der verwendeten Technologie unter statistischen Aspekten ist sowohl für das grundlegende Stoffwechselverständnis als auch für darauf basierende Arbeiten von Bedeutung. Durch den Ansatz des expressionprofiling auf 2-D Gel-Basis werden mehrere Proben miteinander verglichen. Das Ziel dieser Proteomstudien ist die Detektion unterschiedlich exprimierter Stoffwechselproteine. Der Expressionsunterschied zwischen einzelnen Proteinen dient als Hinweis für Variationen im Metabolismus. In Abhängigkeit von den Vergleichsproben und der Analysenmethode wird der detektierte Expressionsunterschied von vielen Parametern beeinflusst. Hierbei muss unter anderem zwischen technischen und biologischen Variationen unterschieden werden (Molloy et al., 2003).
In one specific aspect of this work the potential application of gene expression signatures for the prediction and classification of specific leukemia subtypes was assessed. Today the diagnosis and subclassification of leukemias is based on a controlled application of various techniques including cytomorphology, cytogenetics, fluorescence in situ hybridization, multiparameter flow cytometry, and PCR-based methods. The diagnostic procedure is performed according to a specific algorithm, but is time-consuming, cost- intensive, and requires expert knowledge. Based on a very low number of candidate genes it is demonstrated in this work that prognostically relevant acute leukemia subtypes can be classified using microarray technology. Moreover, in an expanded analysis including 937 patient samples representing 12 distinct clinically relevant acute and chronic leukemia subtypes and healthy, non-leukemia bone marrow specimens a diagnostic prediction accuracy of ~95% was achieved. Thus, given these results it can be postulated that the occurring patterns in gene expression would be so robust that they would allow to predict the leukemia subtype using global gene expressionprofiling technology. This finding is further substantiated through the demonstration that reported differentially expressed genes from the literature, namely pediatric gene expression signatures representing various acute lymphoblastic leukemia (ALL) subtypes, can be used to independently predict the corresponding adult ALL subtypes. Furthermore, it could be demonstrated that microarrays both confirm and reproduce data from standard diagnostic procedures, but also provide very robust results. Parameters such as partial RNA degradation, shipment time of the samples, varying periods of storage of the samples, or target preparations at different time points from either bone marrow or peripheral blood specimens by different operators did not dramatically influence the diagnostic gene expression signatures.
EMMA2 can be locally nistalled, or run via the web interface hosted at Bielefeld University. A user and group management allow to analyze datasets in a group of scientists providing different rights and roles for the data access. The datasets are stored in a relational MySQL database and HDF5 files. EMMA2 uses a LIMS system for raw microarray file storage (ArrayLIMS 3 ). The complete MAGE-OM is used to provide a MAGE-ML compatibility. Various customizable normalization and gene expression analysis pipelines are implemented in EMMA2. A KEGG integration allows to map the gene expression to the KEGG pathway maps and visualize the expression experiments. MAGE-ML, MAGE-TAB[Rayner et al. (2006)] and csv export options are provided by this open source system. None of the three mentioned systems supported one-color microarrays (GeneChips ® ) at the start of the project. As EMMA2 is developed at Bielefeld University and offers the most interesting criteria in the comparison, this project will extends EMMA2
a MLL target gene and upregulated in MLL rearranged leukemia. Based on these interactions DAC treatment may also result in blocking self-renewal of leukemic blasts due to re-expressing genes that are down regulated among these molecular conditions. In summary, DAC therapy seems to overcome certain aggressive leukemia subtype characteristics via different anti- leukemic aspects to, reflecting, thereby rendering DAC an alternative treatment option. Next to the upregulation of genes associated with adverse AML outcome in patients treated with conventional therapy, we also found genes that have been related with favorable outcome including PRAME and INPP5D [104, 149]. The tumor associated antigen (TAA) PRAME functions as a repressor of RAR signaling  and its expression is epigenetically regulated . Just recently a study showed that in leukemia cell lines PRAME impairs the retinoic acid- regulated-cell-proliferation and differentiation, which seemed to be reversed by ATRA leading to a better outcome in patients with NPM1 mutation . The effect of ATRA however does not rely on DNA methylation changes . Compared with our results it seems that AML expressing PRAME shows two ways of therapeutic targeting. One based on reversing the repression of RAR signaling through ATRA and the other one by using PRAME epigenetic based expression and its TAA property by DNA methylation changes and immunomodulation through DAC. Taken together in these types of AML ATRA and DAC may have a synergistic effect leading to an improved response rate or outcome.
Although heavy ion have been applied in clinical therapy of cancers for many years, the genetic mechanisms and the signaling pathways involved in cellular responses to heavy ion radiation are not completely understood. Several previous studies have evaluated the correlation between cellular responses to carbon ion irradiation and the expression status of known genes involved in the regulation of cell cycle, DNA repair, and apoptosis using analytical approach for single gene. Recent studies demonstrated that irradiation with carbon beams induced not only apoptosis, but also cellular senescence in glioma cells with either wild-type or mutant p53 expression, more effectively than X-ray (Guida et al., 2005; Jinno-Oue et al., 2010). Using semiquantitative real time PCR, significant different expressions of 10 selected genes involved in DNA repair have been showed to be responsible to inhibition of potential lethal damage repair in cultured lung cancer cells after carbon ion irradiation compared to X-ray (Yashiro et al., 2007). The expression and focus formation of CDKN1A, a member in the complex of MRE11/RAD50/NBS1 ensuring DSB repair, is correlated with the traversal of ionizing particles (Jakob et al., 2002). Through pathological investigation and immunohistochemical analysis of CDKN1A, carbon ion has been found to be responsible for cell cycle arrest in tumor cells with mitotic catastrophe (Imadome et al., 2008). Recent study using a cDNA expression array containing 161 key genes in damage and repair signaling pathway has revealed that 38 and 24 genes were differentially altered in breast epithelial cell treated with X-ray and heavy ion (Fe +2 ), respectively (Roy et al., 2008).
Microarray gene expression analysis provides a valuable tool to comprehensively examine and identify pathways that are affected by virus infection. Previous transcrip- tional profiling analysis following RV infection of primary human fibroblasts derived from a whole embryo , as well as the ECV304 cell line which exhibit both endothe- lial and epithelial characteristics , revealed that RV in- duces a robust interferon-stimulated gene response. However, since endothelial cells are believed to play a major role in RV-induced teratogenesis, our studies fo- cused on the gene expression changes of an infection caused by a wild type RV isolate (Wuerzburg-12) in pri- mary fetal endothelial cells derived from the human um- bilical vein (HUVEC) and adult endothelial cells derived from the human saphenous vein (HSaVEC). By comparing up- and down-regulated genes in the endothelial cells of fetal and adult origin using Gene ontology (GO) term ana- lysis, we were able to identify differences in biological pro- cesses and pathways between HUVEC and HSaVEC. We believe that these differences in gene expression after in- fection of endothelial cells of adult and fetal origin provide new insights into the molecular mechanisms involved in RV-induced teratogenicity.
Assuming that intron mapping reads originate from nascent mRNAs, zUMIs also counts and collapses intron mapping reads with other reads mapping to the same gene with the same UMI. To assess the information gain from intronic reads to estimate gene expression levels, we analysed a publicly available DroNc- seq mouse brain dataset (, https://portals.broadinstitute. org/single_cell ). For the ⇠ 11, 000 single nuclei of this dataset, the fraction of intron mapping reads of all reads goes upto 61%. Thus, if intronic reads are considered, the mean num- ber of detected genes per cell increases significantly from 1041 for exon reads to 1995 for exon+intron reads (Welch two sam- ple t-test: p-value < 2.2e-16). To assess the impact of intronic reads on the inference of differential expression, we performed power simulations using empirical mean and dispersion dis- tributions from this dataset . The simulations assumed a balanced two-group comparison of variable sample sizes with 10% of the genes differentially expressed between groups. We observed a 0.5% decrease of the marginal false discovery rate (FDR) for exon+intron relative to exon counts for group sam- ple sizes of < 250 cells, while the power to detect differen- tially expressed genes was similar for exon and exon+intron counts. Next, we investigated whether exon+intron count- ing improves the identification of cell types, as suggested in . Following the Seurat pipeline , we clustered the cells of the DroNc-seq dataset based on the exon as well as our exon+intron counts. The KNN-clustering reported 24 distinct clusters for the exon+intron counts, while we could only dis- criminate 15 clusters using exon counts (Figure 4). This analy- sis shows, that the additional genes that were detected by also counting intron-mapping reads are not spurious, but carry bi- ological meaning.
potential functional activity in soil (Torsvik and Øvreas, 2002). Some of the methods to assess potential activity, such as metagenomics, use high-molecular-weight DNA extractions directly from soil to create large-insert libraries of environmental microorganisms (Rondon et al., 2000). Besides, the possibility to detect transcribed mRNA sequences directly in the environmental samples has permitted further insight into the functional activity. Several analytical procedures have been reported, most of them are restricted to isolating mRNA from a) pure cultures (Fleming et al., 1998), b) soil amended with pure cultures (Tsai et al., 1991), and c) soil for targeting specific transcripts by defined RT-PCR assays (Bürgmann et al., 2003; Mendum et al., 1998). Some of the most efficient methods currently available for the comparative analysis of mRNA transcript pools are differential display technique (Fislage et al., 1997; Liang and Pardee, 1992; McClelland and Welsh, 1994; Wong and McClelland, 1994), poly (A) tailing (Grant et al., 2006), and subtractive hybridization (Poretsky et al., 2005). An alternative method for studying mRNA transcripts is to target the transcripts of specific metabolic activity in situ. For example, the in situ hybridization (ISH) of mRNA sequences has been used for studying gene expression in prokaryotic cells (Pernthaler and Amann, 2004) and eukaryotic cells and tissues (John et al., 1969; Gerfen, 1989; Farquharson et al., 1999; Morris et al., 1990; Singer and Ward, 1982). Some of the approaches to quantify microbial gene expression in soil are given in Fig. 5.3.
As assessed by flow cytometry, CD34 + /CD38 − ALL LSCs invariably expressed the stem cell-homing receptors CD44 and CXCR4 (CD184) ( Table 1 and Supplementary Table S5). CD19 was identified on LSCs in patients with B-lineage ALL, but was not detected on LSCs in patients with T ALL or CML. Interestingly, LSCs displayed CALLA (CD10) in patients with Ph + ALL but not in patients with Ph − ALL ( Table 1 ). IL-1RAP, a LSC marker in CML, was found to be expressed on ALL LSCs in a subset of patients (13/ 29 = 45%), including most cases (7/11; 64%) with Ph + ALL (Supplementary Table S5). CD25 was detected on CD34 + /CD38 − ALL LSCs in 12/41 patients (29%), including 11/20 (55%) with Ph + ALL (Supplementary Table S5). CD26 was expressed on ALL LSCs in 5/38 patients (13%), including 5/19 (26%) with Ph + ALL. In Ph − ALL, CD34 + /CD38 − LSCs expressed IL-1RAP in a subset of cases (6/18 = 33%), but displayed CD25 in only 1/21 patients (5%), and did not exhibit CD26 (0/20) ( Table 1 ; Supplementary Table S5). Normal CD34 + /CD38 − BM stem cells did not express CD25, CD26 or IL-1RAP. Since NSG-engrafting LSC also reside in the CD34 + / CD38 + fractions of ALL, we also examined this cell-subset. In patients with Ph + ALL, these cells expressed CD10, CD44, CD184 and IL-1RAP (Supplementary Tables S6 and S7). Overall, the phenotype of CD34 + /CD38 + stem/progenitor cells in our ALL Table 1. Expression of Lineage-related Markers, Stem Cell Markers, and Niche-related Antigens on ALL LSCs, CML LSCs, and Normal CD34 + /CD38 − Bone Marrow (BM) Stem Cells
Cell motility is crucial for the effective function of T lymphocyte and immune homeostasis. Priming of lymphocytes for adaptive immune responses requires their migration from the blood stream to lymph nodes to allow their interaction with antigen loaded APCs. Distinct lymphocyte subsets use different combinations of cell migration factors on their surface to orchestrate effector functions and for general immune surveillance. Lymphocyte homing to skin requires the expression of selectin E (SELE) and selectin P (SELP) ligands, CC-chemokine receptor 4 (CCR4) and CC-chemokine receptor 10 (CCR10) (45). Likewise, lymphocyte tropism for the intestines is controlled by expression of the intestine homing receptor α4β7 and CC-chemokine receptor 9 (CCR9) which binds to mucosal addressin cell adhesion molecule 1 (MAdCAM1) and CC-chemokine ligand 25 (CCL25), respectively (46). Up to now, no adhesion molecule combination has been defined that specifically target (effector) T cells to the CNS.
Ziel dieser Arbeit war es, bisher unbekannte, früher Zielgene des CALM/AF10-Fusionsproteins zu identifizieren. Dazu wurde das Fusions-Gen in den episomalen pRTS-1 Vektor einkloniert, dessen Promotor mit Hilfe des Tetrazyklin-Derivats Doxyzyklin conditional aktivierbar ist (Tet-on System). Dieses Kontrukt wurde dann in die Burkitt-Lymphomzelllinie DG75 stabil transfiziert und die Expression von CALM/AF10 nach Induktion des Vektors mittels RT-PCR überprüft. Als Negativkontrolle dienten DG75 Zellen, die mit einem pRTS-1 Vektor ohne Fusionsgen stabil transfiziert worden waren. Mit Hilfe des Genchips Affymetrix® hgU133 plus 2.0 wurden Genexpressionsprofile erstellt. Um zwischen frühen und späten effekten der CALM/AF10 Expression unterscheiden zu können, wurden Genchipanalysen zu unterschiedlichen Zeitpunkten (24h und 72h nach Vektorinduktion) durchgeführt.
glycerol and FA, were upregulated. In addition, the expression of genes involved in FA metabolism, including acyl-CoA thioesterase (Acot1), were enhanced up to 90- fold. Expressions of the PPARα target genes CYP4A11 and FA binding protein (FABP1) were induced, and are involved in catabolism and intracellular transport of FA. Genes involved in the import of FA into the mitochondria, like carnitine O- octanoyltransferase (CROT), carnitine O-acetyltransferase (CRAT) and carnitine palmitoyltransferase (CPT1B), were also induced as well as several genes involved in mitochondrial and peroxisomal FA catabolism. Interestingly, the strongest upregulations were observed on day three (Appendix Table 7.7), which correlates well with the gene expression study performed in livers of rats treated with EMD [Ellinger-Ziegelbauer 2011]. Cholesterol synthesis seemed to be downregulated since genes encoding lecithin-cholesterol acyltransferase (LCAT) and ATP citrate lyase (ACLY) were repressed while genes involved in fat storage, such as fat storage-inducing transmembrane protein (FITM2) and perilipin (PLIN2), were induced. Genes associated with peroxisome proliferation, a typical effect of PPARα agonists in rodents [Li 2002a], were upregulated up to 5.1-fold. These included genes involved in peroxisomal FA catabolism, for example acyl-CoA thioesterase (ACOT8) and the PPARα target gene acyl-CoA oxidase (ACOX1), as well as genes related to peroxisomal biogenesis and division (e.g., peroxisomal biogenesis factors 11 alpha and 19 (PEX11A, PEX19)). The expression changes of genes involved in lipid metabolism caused by EMD is consistent with gene expression results of hepatocytes and livers of rats treated with known PPARα agonists [Guo 2006, Tamura 2006] as well as with the gene expression profile of the FF treated cells reported in this study. Thus, these findings support the hypothesis of EMD being a potential PPARα agonist. However, the PPARα gene itself was not upregulated by EMD during the present study, but this is in agreement with previously published reports for FF and other PPARα agonists [Guo 2006, Tamura 2006, Ellinger- Ziegelbauer 2011, Sposny 2011].
weight, food and water intake, and spout preference (Bachmanov et al., 2002). In this atudy it was also shown that AKR/J mice were heavier and ate more (diet g/mouse) than SWR/J mice. However, if the food intake was adjusted by body weight, SWR/J mice ate more (diet g/body weight g) (Bachmanov et al., 2002). In other studies with respect to the preference of macronutrient diet selection, the lean strain of SWR/J consumed more calories from carbohydrate diet whereas AKR/J consumed more calories in form of fat (Smith et al., 1997;Smith et al., 1999;Smith et al., 2000;Smith et al., 2001). The sensitivity of dietary obesity was reported by West et al., when exposed to high fat diet, AKR/J mice consumed more energy and had more fat content (West et al., 1992;West et al., 1995). For these two inbred mouse strains – AKR/J (diet-induce obesity model, DIO) and SWR/J (diet-resistant model, DR), although many dietary studies were reported, only Prpic et al. investigated strain specific differences in the gene expression of uncoupling protein (UCP) 1 and 2 in adipocytes during diet-induced obesity (Prpic et al., 2002). They reported that HF diet induced a modest increase in brown adipose tissue (BAT) UCP1 mRNA in SWR/J mice, whereas a large decrease in UCP1 expression in AKR/J mice, and that UCP2 was consistently higher in white adipose tissue (WAT) from AKR/J than in SWR/J mice and induced by the HF diet in AKR/J but not SWR/J mice (Prpic et al., 2002).
Lakshmipathy et al., 2007; Enerly et al., 2011 ), apoptosis ( Hwang and Mendell, 2006 ) and tumorigenesis ( Ventura and Jacks, 2009 ). MiRNAs belong to the class of small RNAs, which are non-coding RNAs with a length of less than 300 nt. Small nucleolar RNAs (snoRNAs), small nuclear RNAs (snRNAs), short interfering RNAs (siRNAs) and Piwi-interacting RNAs (piRNAs) also belong to this class. Until recently, small RNAs were regarded as evolutionary trash and RNAs that do not encode protein are often considered as results of “leakage” of the transcription machinery. However, present researches have highlighted that ncRNAs can have a wide range of functions and can be divided into different classes by sizes or functions (Figure 4). The regulatory functions of small RNAs in gene expression and their pivotal roles in physiological as well as pathological processes revealed them to be vital components of the genome ( Kutter and Svoboda, 2008 ).
Johansson et al. used the concept of “call blocks” to find the best time instant to inject a fault into Windows CE. Net DDs [Johansson et al., 2007a]. Call blocks were obtained on the fly by monitoring the communication inter- face of the selected DD with the rest of the OS kernel. They could be also considered a form of an operational profile, as they represent runtime behav- ior in terms of sequences of calls to driver-external functions. Instead of using the more common approach of triggering an error injection on the first call of an external function, Johansson et al. explore the effects of triggering injec- tions on a call block basis. The operational profiling methodology presented in the Chapter 7 of this thesis is similar in the sense that operational profiles are built using an equivalent source of information. Though, the approach presented in Chapter 7 is developed to reveal execution hotspots in DDs in terms of followed code paths.