====== Worked example Differential signaling ====== ===== The oestrogen receptor (ER+) positive group Vs negative group (ER-) ===== ==== Differential signaling inputs ==== **Note:** You can also launch the Differential signaling example by clicking the //Run example// button or the //help// button on the header and then //Differential signaling example//. **1-** Log into HiPathia. For further information on this step visit [[logging_in|logging in]]. **2-** Collection of data. We will work with a Breast Cancer dataset from the repository The Cancer Genome Atlas (TCGA) [[https://portal.gdc.cancer.gov/projects/TCGA-BRCA | Link to dataset]].\\ More information on the proposed dataset is available here: * https://www.nature.com/articles/nature11412 * https://pubmed.ncbi.nlm.nih.gov/23644459/ Before use in HiPathia, the dataset must be normalized. We recommend using the [[https://genomebiology.biomedcentral.com/articles/10.1186/gb-2010-11-3-r25 | logarithm of the trimmed mean of M values]] (log2TMM). We have randomly selected 178 samples of breast cancer tumors from the dataset, annotated as ER+ and ER-(the molecular annotations come from [[https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3465532/ | this paper]]). This tumor classification considers the presence or absence, respectively, of estrogen receptors in tumor cells. The goal of this study would be to learn how different these cancer subtypes are at the signaling level, and what processes are differentially active between them. The expression matrix and the experimental design can be downloaded from these links: * Expression matrix: [[http://hipathia.babelomics.org/data/brca_data_example_ERposneg.tsv|brca_data_example_ERposneg.tsv]] * Experimental design:[[http://hipathia.babelomics.org/data/brca_designmatrix_ERposneg.tsv|brca_designmatrix_ERposneg.tsv]] **3-** Upload the normalized data to HiPathia by clicking //My data// in the data panel, or click the //run example// button. For further information on this step visit [[upload_your_data|Upload your data]]. **4-** Click //Differential signaling// button. {{ :hipathia_bar_diffsig_6.png?600 |}} **5-** In the //Input data// panel, click the //File browser// button of the section //Expression matrix// and select the desired normalized expression file. {{ :hipathia_work1_6.png?600 |}} As mentioned before, the number of samples is 356 (178 ER+ and 178 ER-), and this information appears after selecting the expression matrix file: {{ :hipathia_work1-2_6.png?600 |}} **6-** In the //Design data// panel, select //Two group comparison//. Click //File browser// in the //Experimental design// section, and select the desired file. {{ :hipathia_work2_6.png?600 |}} Automatically, //Condition 1// and //Condition 2// files are selected. Change //Condition 1// to "Negative" and //Condition 2// to "Positive" for it tobe used as the reference condition. {{ :hipathia_work2-2_6.png?600 |}} **7-** Select //Human (Homo sapiens)// as species (default). {{ ::examplespecies.png?nolink |}} **8-** In //Effector annotation source//, check both the //Gene ontology// and //Uniprot keywords// boxes. **9-** In the //Pathways// panel select all the pathways (all are selected by default). **10-** In the //Study information// panel, click the //File browser// button and select the desired output folder. In this case we will use //analysis_BRCA//. {{ :hipathia_work3_6.png?600 |}} Give a name to the study, for example, "Differential signaling: Er+ Vs Er-". {{ :hipathia_work3-2_6.png?600 |}} **11-** Click the //Run analysis// button. {{ :hipathia_work1_all.png?600 |}} A study will be created and listed in the studies panel. You can access this panel by clicking the //My studies// button. {{ :hipathia_work1_study-browser.png?600 |}} This a video tutorial for differential signaling study: {{:youtube.png?direct&30|https://www.youtube.com/watch?v=ytyagfQpQPk&feature=youtu.be}}:https://www.youtube.com/watch?v=ytyagfQpQPk&feature=youtu.be ==== Differential signaling results and interpretations ==== Once the analysis is finished at HiPathia web, the report/results will be available in “My studies”.\\ The results page of the Differential signaling tool includes different output results. You can download any table or image showed in the report page by clicking the name right before it.\\ You can also download the circuit activity values and function matrices by clicking //Circuit values//, //GO terms values// and //Uniprot keywords values// respectively. For further information about the differential signaling report please visit [[differential_signaling#differential_signaling_report|Differential signaling results]] /* Furthermore, the report can be downloaded to be visualized locally. {{ ::downloadreport.png?nolink |}} The downloaded RAR file has to be extracted then you can open the //"index.html"// file using a web browser like Firefox web browser, some configuration may be needed for chrome or other web browsers: installation of web server( for example:[[http://www.wampserver.com| Wamp]] or [[https://www.apachefriends.org/fr/index.html|XAMPP]]) then open the report from the navigator using http://localhost/Path/to/my/downloaded_report/) The results are divided in different panels: === Study Information === There is more information about the study available here: {{ ::studyinforeport_6.png?nolink |}} * **Name**: the study name. * **Description**: the description of the current study. * **Tool**: the name of the used tool (in this case, is Hipathia). * **Date**: the launching date of the study (MM/DD/AAAA, HH:MM:SS AM/PM format) === Input parameters === Here you can visualize the parameters with which the current study was launched. {{ ::iputparamreport_6.png?nolink |}} * **Expression file**: The name of the expression file that has been used in the current study. * **Design file**: The name of the design file that has been used in the current study. * **Comparison**: The groups that have been compared, for example; Normal vs Tumor. * **Paired analysis**: Have the input data been paired? **No** or **Yes**. * **Species**: The species of this experiment; Human (//Homo sapiens//), Mouse (//Mus musculus//) or Rat (//Rattus norvegicus//). === Pathways === {{ :pathwaysreport_6.png?nolink |}} A pathway viewer has been developed to facilitate the visualization of the comparison results. It shows a graphical representation of the activation levels of the "Effector circuit" comparison between the two groups (ER+ and ER-). The meaning of the different symbols that can be found are in the legend below. {{ ::report_legend.png?nolink |}} The information summarized in the pathway viewer includes: Node shape: * **Ellipses** correspond to genes. * **Circles** represent metabolites. * **Rectangles** are cellular functions. Node color: * **Blue** for down-regulated genes, **red** for up-regulated genes and **grey** for non-significantly differentially expressed genes. * The intensity of the color depends on the significance of the differential expression. Edge shape: Edges represent either activations or inhibitions. * **Activations** are represented by classic arrows. * **Inhibitions** are represented by T arrows. Edge color: * Circuits which are **up-regulated** in the ER- condition have their arrows colored in **red**. * Circuits which are **down-regulated** in the ER- condition (condition 1 ) have their arrows colored in **blue**. * Circuits which are **not significant** for the test have their arrows colored in **grey**. * **When an edge belongs to different circuits which should be in different colors, different arrows are depicted, each one for each necessary color.** The pathway viewer has four principal parts in order to facilitate the visualization of the results: {{ :pathwaysreport1_6.png?nolink|}} \\ In the upper-right part of the visualization tool, all selected pathways from the differential signaling form are shown along with one or two arrows. These arrows indicate whether in one of the "Effector circuit" within each pathway a differential activation pattern between the two compared groups have been found.\\ The arrows will be colored if the differential activation is significant after the p-value adjustment (or unadjusted p-value, if the "Unadjusted" parameter has been selected) and depicted in grey if it isn't.\\ The arrow will point up or down if an up-regulation or down-regulation of the signal occurs between the circuits within that pathway.\\ The meaning of this up or down-regulation depends on the comparison performed, that normally would be case vs. control (comparison between selected class from the design file : condition 1 Vs condition 2).\\ Only one arrow is shown if all the effector circuits are whether up or down-regulated.\\ ---- {{ :pathwaysreport2_6.png?nolink|}} In the lower-right part of the tool will appear all the circuits in which a pathway (previously selected on the upper part -1-) can be decomposed.\\ Once any of the circuit is clicked the nodes and interactions (edges) that form part of this circuit are highlighted in the pathway viewer (), One example might be the red-highlighted circuit on the figure below. {{ ::effectorcircuithl_6.png?nolink| }} ---- In the visualization part there are two types of objects, the nodes and the edges. {{ :pathwaysreport3_6.png?nolink |}} * The nodes represent the different proteins or metabolites that are responsible on passing the signal from the previous node to the next one. As commented before, the nodes can be plain or complexes. In the former it is only necessary the presence of one protein (although that same function can be done by several proteins), while in the complex ones it is necessary the presence of two or more proteins to pass the signal through.\\ The color on the nodes are the result of a differential expression analysis on the gene/s involved on performing that node's function. * The edges represent how the interactions between the different nodes are.\\ If the edge is an arrow then the previous node will be activating the next one, while if it ends with a vertical bar is the former node will inhibit the functionality of the following node. This interactions may be depicted in red or blue depending on the circuit they form part of (whether they are up-regulated or down-regulated). It may occur, that two or even three colors for a same edge are shown, but this only happens when a circuit is not yet selected on the lower-right part of the tool.\\ Once an certain circuit is selected all its edges will be colored in the same color depending on the result of the differential signaling activation analysis.\\ ---- The top part contains the title of selected pathway,by clicking this button {{::keggsource.png?nolink|}} you can see the original source of this pathway. {{ :pathwaysreport4_6.png?nolink |}} You can also find other buttons like: * {{::search.png?nolink|}} to search specific genes, proteins or functions. {{ ::searchgif.gif?nolink |}} * {{::export.png?nolink|}} to export a SVG image of viewed objects (the whole pathway or just the selected effector circuit) * {{::center.png?nolink|}} to center the selected pathway * {{::width.png?nolink|}} to adjust width * {{::height.png?nolink|}} to adjust height === Circuit values === Here you can find additional tables and plots: * The first section below contains a link to download the computed circuit activity values. This matrix file indicates for each effector circuit the level of activation calculated using the HiPathia method for each sample.{{ ::circuitvalue.png?nolink |}} * The Heatmap plot represented on the hipathia results page is a heatmap of the activation values from the most differentiated effector circuits (rows) between groups along with a clustering of the samples (columns). This plot allows to observe if its possible to differentiate the groups that are compared according to its effector circuit activation values.\\ The colors depicted here indicates the level of activation for the different circuit in each sample, being the bluish ones those with lower activation levels and the reddish the highest ones (e.g. if a given cell of the heatmap is blue would indicate that this particular effector circuit is poorly activated in that determined sample) {{ ::paths_heatmap.png?nolink |}} * The next plot on the report corresponds to a Principal Component Analysis plot. This figure is useful to determine if the activation levels of the pathways calculated with Hipathia are able to differentiate between the two groups that are being compared. {{ ::paths_pca.png?nolink |}} * In the section below there is a table (that can be downloaded) with the results of the differential activation analysis. This table indicates for each effector circuit whether or not there is a different level of activation depending on the group the samples belong. {{ ::circuitsig_6.png?nolink |}} * **circuit/term**: Code identifying the circuit consisting in the name of the pathway to which it belongs and the effector (last) node of the Circuit (//pathway Name : Effector//). * **UP/DOWN**: Up when the circuit is up-regulated in the disease case, DOWN otherwise. * **statistic**: Statistic of the Wilcoxon test performed between conditions. * **p.value**: P-value of the statistic test. * **FDRp.value**: Adjusted p-value of the statistic test using the FDR method. * **Fold Change**: is a measure describing how much the ** circuit activity** changes between conditions.[[https://en.wikipedia.org/wiki/Fold_change|reed more about Fold Change]] * **logFC**: The logarithm to base 2 of calculated Fold change. === Function based analysis === For each of the functional analysis selected to be performed, the same descriptive statistics as in the Path values case are shown: * **GO terms/Uniprot keywords values**: You can download the matrix of function values by clicking //GO terms values// or //Uniprot keywords values//. * **Heatmap**: A heatmap is a graphical representation of the values of a matrix. A not-supervised clustering is performed on the data per columns (samples). A good separation between the two colors of the upper bar of the heatmap indicates that the differences between the two groups are big enough to discriminate between them. * **PCA**: A Principal Components Analysis is performed in order to see if the matrix contains information enough to separate between the two groups. * **GO terms/Uniprot keywords significance**: Ordered table of the functions significance. More significant terms are displayed in the upper part of the table. For each functional term, the following information is shown: * **path/term**: Name of the function. * **UP/DOWN**: Up when the function is up-regulated in the disease case, DOWN otherwise. * **statistic**: Statistic of the of the statistic test. * **p.value**: P-value of the statistic test. * **FDRp.value**: Adjusted p-value of the statistic test using the FDR method. * **Fold Change**: is a measure describing how much the ** circuit activity** changes between conditions.[[https://en.wikipedia.org/wiki/Fold_change|reed more about Fold Change]] * **logFC**: The logarithm to base 2 of calculated Fold change. */ /* ==== Results interpretations ==== */ **Estrogen receptor (ER) pathway (hsa04915):** If we look at the estrogen receptor pathway, we can see it is mainly downregulated in ER-negative (ER-) tumors. Considering that ER- tumors are characterized by lacking estrogen receptors, these results are consistent with prior knowledge that the estrogen pathway is downregulated in tumors that are not sensitive to estrogen [1]. {{ :estrogen_signaling_pathway_hsa04915_.jpg?nolink |}} Despite gene expression being upregulated in ER- for CREB3, HiPathia shows its activity to be decreased according to the signaling cascade that activates it. The role of CREB3 in cancer progression and metastasis is controversial, being considered both a marker of good and bad prognosis, thus the interpretation of these findings might be tricky. The involvement of ER related CREB3 in breast cancer seems to be related to the activation of the transcription of other bad prognosis genes, such as CXCR4, ARF4, USO1, among others [2][3], thus the interpretation should be done along with the information of other pathways. Nevertheless, and despite the initial thought that ER- tumors will have all CREB3 circuits downregulated, in this Hipathia simulation we have found a mechanism that allows ER- cells to activate CREB3, and therefore, to putatively promote migration and metastasis, and that deserves further research. **Focal adhesion pathway (hsa04510):** {{ :focal_adhesion_hsa04510_.jpg?nolink |}} In the focal adhesion pathway, the activity of proteins involved in actin polymerization, such as actinin, vinculin, and zyxin, is upregulated in ER- tumors. Actin polymerization is involved in cell migration processes required for metastasis, which explains their upregulation given that ER- tumors are known to be more prone to migration and metastasis [4]. ==== References ==== [1] Frasor J, Danes JM, Komm B, Chang KC, Lyttle CR, Katzenellenbogen BS. Profiling of estrogen up- and down-regulated gene expression in human breast cancer cells: insights into gene networks and pathways underlying estrogenic control of proliferation and cell phenotype. Endocrinology. 2003 Oct;144(10):4562-74. DOI: 10.1210/en.2003-0567. Epub 2003 Jul 10. PMID: 12959972. [2] Howley, B. V., Link, L. A., Grelet, S., El-Sabban, M., & Howe, P. H. (2018). A CREB3-regulated ER-Golgi trafficking signature promotes metastatic progression in breast cancer. Oncogene, 37(10), 1308–1325. DOI: 10.1038/s41388-017-0023-0. [3] Kim, HC., Choi, KC., Choi, HK. et al. HDAC3 selectively represses CREB3-mediated transcription and migration of metastatic breast cancer cells. Cell. Mol. Life Sci. 67, 3499–3510 (2010). DOI: 10.1007/s00018-010-0388-5. [4] Padilla-Rodriguez, M., Parker, S.S., Adams, D.G. et al. The actin cytoskeletal architecture of estrogen receptor positive breast cancer cells suppresses invasion. Nat Commun 9, 2980 (2018). DOI: 10.1038/s41467-018-05367-2.