User`s Manual - University of Michigan
Contents
1. Diabetic Neuropathy Microarray Knowledge Base DNMKB User s Manual Released on 1 2 2014 http jdrf neurology med umich edu DNMKB Copyright 2013 Program for Neurology Research and Discovery All rights reserved Feldman Laboratory Program for Neurology Research and Discovery University of Michigan Ann Arbor MI 48109 USA Email DNMKB help umich edu Table of Contents Introducing DNMK B iusssseckecsktu bud ue nera d uB E ud SYM MUR ecl ea caEdS KERN PCUM UE 1 See e EEE EE E E E E 3 ACCES MIO ITS e E E 4 EDI 4 jnre c UN 6 ofer TOT 6 Jeler ee esee EEG MEME UM EPUM E 6 MaN MENU ct 7 BROWS CIV AIV NNUS UU TETTTTTUTMTM 7 TS I Ea E MERE O T ETT 8 ANAN SIS NIO e E E RR 9 Unde rstanding lt T 11 Introducing DNMKB Diabetic neuropathy DN is the most common and debilitating complication of diabetes but the pathogenesis is not fully understood despite extensive research Recently the DN research community employed high throughput technologies to examine DN associated transcriptomic changes in human and animal models To comprehensively understand the complex systems associated with DN it is critical to have a disease specific data storage and analysis system to facilitate effective mining and seamless integration of the enormous amount of data Here we present the Diabetic Neuropathy Microarray Knowledge Base DNMKB a centralized repository and analysis portal of diabe2. Ontology term and KEGG pathway name or their respective ID and explore gene regulation across studies All the results are downloadable in Excel spread sheets to facilitate the users downstream analyses Browse mode Users can browse the compendia of the transcriptomics datasets The full list of differentially expressed genes DEGs will be available for any selected dataset using user defined significance level and fold change criteria Search mode Users can search the database using various criteria such as gene name Entrez ID gene symbol or synonym and associated biological function Gene Ontology term and KEGG pathway name or their respective ID Analysis mode Users can perform various analyses on the DEG sets Currently supported features include functional enrichment analysis for identifying enriched biological functions among the DEGs gene set analysis for identifying the gene level overlap among selected DEG sets and transcriptional network analysis for network level comparison of two selected DEG sets Starting DNMKB Login Click the LOGIN button on the front page of DNMKB and proceed with registered or public ID and password Public users can enter public access as email and public as password or simply click the Public Login button If you are a registered user enter your email address and password Public users can use public access as email and public as password If
3. dbdb Chorttxt Gbdb DRG HwK db vs dbdb Chart ads gah tt dbdb DRG Significance heatMap up ta top 10 terms per set Chart full Annotation Heat map information amp S S a eer 3 fee gaa HT 44 ais Sag rie x 79 m Tu d 2 6 EE K xL d ata E LES vI 31h BA i m Wd d quae gue E L oe m gea Nagra lebe T aN L K L L Top 10 most over Hal represented biological ai p M functions scien Adar enar VETT Ber a TETA Aer E H1 cuni y digan proc Greed el CS es ation LI mmeona Ac rob GPCRS Poder d dT m N Cx ada T ie radieuam Put Hnenimag color 6 based an ig 1 Py aloe A ble cell amples no sagnifieanl enrichment Al pilher celis wih pinkirmd color indicabe slalisBisal significance Pao eR The more red the meen significa Download the complete results FuncEnrichment zip Figure 7 Functional Enrichment Analysis Results Gene Set Analysis GSA The result of GSA is a Venn diagram showing the overlap between DEG sets DNMKB also provides the list of overlapping genes available for download IE Diabetic Neuropathy Microarray Knowledge Base Geneset overlap Analysis CHB Lipp araa 5 24 TSP va P E dI RHO Re KE va ded The name of DEG set The number of intersection genes between DEG sets Summary j a Py By TA ES Rim MEE j i la 74 b AND ORI wk didi ANE dhdi pai diraeh wy E AHO Pons DAG
4. imat va dbdb oniy dbdb Hisecox meus 236706 ve dbo DAG 2405 db vs deda AND shan ORS Bas dis vs dbdbdb a l Shab Hippoacam 4 hostis pp Sw us 1335 dbl only 183 de DRG Bak be adhd 558 dixib Hippoocampui de dta t dhdh 12 dedib 64 Sub gene sets Bulc DRG ahak ire ws Sib OG ave db ys bdb dhdbh DRG Gwi gia bdb DROo fawkes Abb ORG Her ee v5 abdb ngGo z4wk dee vs A ire dbdb mippocampus wk 5 0 vti dhidi didl Hippacampus Gladis DEI Bwb db vs dili dbdh dixib Hgipucamgpuss2 Summary Ghat ie v5 ghat dbdb dbdb Huppocampusz Awk db vs dih AEZ dR vs diio vp Gil Awha i v dls Figure 8 Gene Set Analysis Results Transcriptional Network Analysis TNA TAN will be performed between two selected DEG sets DNMKB will use 6 levels of allowed mismatch a k a approximate parameter in TALE ranging from 0 perfect match to 0 5 allowing up to 5096 mismatch Summary of the shared sub networks is given at the bottom of the result page along with links to the network format file gml which can be loaded into Cytoscape hitp www cytoscape org an open source network visualization tool A simple image will be generated for those networks with less than 300 nodes The list of genes as entrez gene IDs will be available for download as well 14 DNMKB Co Citation Network Analysis Congratulation Your co citation network analysis has been successfully completed Your selected differenti
5. you forgot your password contact your system administrator E mail Password Public Login Figure 1 Login Select Options Diabetic Neuropathy Microarray Knowledge Base Instruction Choose options and or keyword to retrive microarray data after then click Retrieve Available Microarray List button for microarray results Or click Reset to Default button to reset all selected options to their default values This step is optional Just click the Retrieve Available Microarray List button to see all available data sets Species Tissue Age DEG Tool k ud v Yv K 4 Keyword Retrieve Available Microarray List Reset to Defaults Figure 2 Select options The first step after login is to retrieve the available microarray data sets and filter them using four major criteria and or simple keywords e Drop down menu Select options for Species Tissue Age and DEG tool or type search keywords into the textbox to retrieve microarray dataset Then click Retrieve Available Microarray List button It should be noted that this filtering step is optional but users must click the retrieve button to proceed e Reset Click Reset to Defaults button to reset all selected options to their default values Main Menu Browse Menu BROWSE provides users with an efficient way to retrieve all DEGs with their fold change information The results can be so
6. ally expressed Detai E of selected sets were DEG sets HIDE analysis option summary DNMKB exp ID 39 Larger DEG set Number of DEGs 4017 Name dbdb SCN 24wk db vs dbdb txt DNMKB exp ID 5 Number of DEGs 4938 Shared sub networks Network file gml can be identified by TALE loaded onto Cytoscape HumanDN Cl Sural NA NonProg vs Prog txt Shared sub networks Network file server for a can be loaded into Cytoscape a open source network visualization tc Shared Stwork with over 300 nodes gene lists Network of allowed of seed of nodes in of nodes in Network file number mismatch nodes shared network shared network gml Download network en l SD Genes 10 152 2489 Genes file qml Download network AA 10 404 5591 jap Vig teste Genes file qml 10 595 8774 janie tovians Genes file qml 10 Download network Network is 742 7200 too big tn chow Genes file qm 3 20 4 20 5 40 eee Download network hiii 797 i n file cami too big tn show Genes END OF THE USER S MANUAL 15
7. an FSS ORO ae y Ce org Ibm 0 059 Set va EOG in Eh DH a ui un Miohigpan C e e i e K ieee Figure 4 Search Menu Des play Order Experiment name i Species Tissue Ageqwks DEG Tool Couto Hote PubMed IU Lab FI Published Analysis Menu ANALYSIS provides users with further analysis tools to identify meaningful information from selected DEG sets Three different analysis methods are currently supported in DNMKB Functional Enrichment Analysis Gene Set Analysis and Transcriptional Network Analysis Diabetic Neuropathy Microarray Knowledge Base Functional Enrichment Analysts Cellular Component Gane Set Analysis Transcriptional Network Analysis Biological Process Molecular Function KEGG Pathway Generate Venn Diagram Generate Co Citation Network Generate Heat Map Significance Cut off lt Minimum Fald Change List of available Micrearray Experimenta i Toggle Dregday Codes Experiment mame E Leach Tissue Agetks DED Tea Culofl PubMed ID Publishi is E M E M K M R E M RE KR E R E E E R M OA Figure 5 Analysis menu Functional Enrichment Analysis FEA Gene Ontology GO http Awww geneontology org terms and Kyoto Encyclopedia of Genes and Genomes KEGG http www genome jp kegg pathway information are used in FEA Basically GO is classified by three groups Cellula
8. atrix Browse Results Download the matrix in Excel file This links to NC ne records Cyclin dependent kinase mhibitoi 1A P21 esterase Ltormylglutathig ne hydrolase Bemardinelli Seip congenital lipodystrophy 2 homolog human pois sium voltage gated channel shaker Clicking column header will sort the table dbdb_24w k SCN dbdb 24w k Hippoca mpus dbdb 8wk Human Pr Hippoca ogressive_ mpus DN Sural 421 1 12 mbol2 Count Berardinelli Seip congenital lipodystrophy 2 DEG fold changes related subfamily Unless specified otherwise the comparison is between Clicking symbol will create a DI uncti ary of this genes solute cai family 7 cationic amino acid Eee e The degree of fold change is also represented by aai l color gradient of the cell red vs blue mE Figure 6 Browse Menu Result Table 55301 SLCTATU t e The matrix is downloadable in Excel file e Clicking the column headers will sort the table e Clicking gene IDs will show the detailed gene information NCBI Entrez Gene database e Clicking symbols will create a summary page of biological functions GO and KEGG pathway associated with the selected gene e The values correspond to the fold changes between control and diabetes unless specified otherwise Positive values up regulated in diabetes and negative values down regulated in diabetes The degree of fold chan
9. ge is also represented by color gradient of the cell red vs blue As shown above clicking symbols will create a summary page for biological functions in terms of GO and KEGG pathway associated with the selected gene in a new window Depending on the number of associated function the loading time of this page may take 11 up to a minute So be patient The current DNMKB displays not only those explicitly assigned GO terms but also those implicitly assigned GO terms as well which can be inferred from the explicitly assigned GO terms and the hierarchical GO structure Future version will allow users to select which sets of GO terms to use explicitly assigned terms are less in number thus taking much less time to load The number Heer Shen to ther of D in the selected datasets e i z i z Iadalallalezal Ic OF aS Sino o L LIII TU EL L 42 C H L a HL TU GC d First number associated DEGs in all the selected dataset no orthologous genes combined Second number total number of genes associated with the selected GO term human mouse rat Figure 6 Summary Table for Biological Functions Search Results The SEARCH result table is similar to the BROWSE menu except that in the SEARCH menu users can specifically search for the DEGs in the database using various criteria Analysis Results Functional Enrichment Analysis The results of FEA are gene annotation information of the enrichmen
10. he drop down menu is different based on the selected datasets resulting by the previous step Combine multiple species This option specifies how genes from multiple species are handled and displayed The default is use mouse gene as base as the majority of the datasets are using mouse Search Menu SEARCH provides users with search flexibilities to retrieve specific DEGs of interest As in Browse menu users can provide custom significance and fold change cutoffs If these values are not specified the default values for each DEG set will be used Once the keywords or significance and fold change cutoff values are typed users click the Generate Matrix button Search criteria DNMKB supports seven types of search criteria Only one search criterion should be used for each query although multiple keywords are allowed in selected criteria noted as MULTI below Allowed separators include semicolon comma tab space newline Gene IDs Entrez gene IDs MULTI Gene Symbols Entrez gene symbols either official or synonyms MULTI Gene Names Entrez gene name either complete or partial names GO IDs Gene Ontology IDs MULTI GO Term Gene Ontology term either complete or partial terms KEGG IDs KEGG Pathway IDs MULTI KEGG Pathway Name KEGG pathway either complete or partial names Show non DEGs option This option specifies if the result matrix will include any non DEGs Th
11. is feature is useful in case the genes of users interests do not show up in the matrix and users want to make sure if the genes are included in the array platform The default is Do NOT show any non DEGs If Show any non DEGs is selected the following colors will be used to represent different DEG types e Green included in the array and a DEG e White included in the array but not a DEG e Gray not included in the array All other features in the SEARCH menu are identical to those in the BROWSE menu Diabetic Neuropathy Microarray Knowledge Base Search by gene symbol name ID GO KEGG pathways accepted separators semicolon comma tab space newline tor multiple termms105 IDE dedaied search options Senrch by termalDa Sere Hame Hampl supersxide dritt i DJ 179 jem pli LOO D 15 CTET p oO Tem mip aixidatbee Hess CS he oak recede amp aMITICH Fadl T KEGG Pray Hama example diabetes mHlitur amp how non DEGs Da MOT show amy non DEGs Sort by of expenmenits with DEGs 7 Combine multiple species USS moute ea 35 base T Geman Mat Reset Hote Combining multiple species works with two species Significance Cut off lt Minimum Feld Change List of available Microarray Experiments teret T puisi NS PSU 5 X e I3 EU or 1 2i I me DAG at IBMIT 0 080 m dada In a DAS 2 Unounahed 1 Michig
12. password are given by the system administrator and non members can use the following login Email public access Password public Table 2 Summary of current transcriptomics data in DNMKB of Data Set Name Species Type Genotype BG Age Tissue DEG Treatment Published sets de mose 2 BKSd b s24ws sonore f w a bldoauononic mouse 2 Bed we 0G i No L wm m ams mw stone 2 0 Do me 4 pg Se SN 3 Rosigitazone Yes ro mm sa mots Sus soon a_ rane to _ obbmde mouse 2 BBRoob 3ws SN 8 m ebmde mouse 2 BIB bb zws SN 0 No so mee na cemu 23m son 7 ves Hunanon mn igo m M sa 2 w iRa CH 1 Gookakzaki e Hippocampus Pat 2 ia Pozna t afa 69s DG 3 a Lmem Ww enm m C2 vos Puteo mouse 4 mo actar RRR sonore 2 Ye DRG dorsal root ganglia SCN sciatic nerve AG autonomic pa Features DNMKB allows users to explore the compendia of genes and biological functions pathways perturbed in the neuronal tissue by diabetes or drug treatment Users can easily identify the most frequently and highly regulated genes in either all or selected datasets across different animal models tissues and ages Users can search the database using various criteria such as gene name Entrez ID gene symbol or synonym and associated biological function Gene
13. r Component Biological Process and Molecular Function Therefore DNMKB provides four different categories to perform the functional enrichment analysis Users can select as many DEG sets as they want and then click the Generate Matrix button FEA generates heat maps for selected DEG sets Users can click heat map icons to see the bigger images for the maps If users want to download all the information generated by FEA click Download the complete results FuncEnrichment zip Gene Set Analysis GSA GSA is done by clicking the Generate Venn Diagram button after selecting between two and 5 DEG sets GSA provides a Venn diagram showing the number of overlapped gene sets between them Transcriptional Network Analysis TNA TNA identifies conserved transcriptional networks between two DEG sets based on the gene gene co citation data Sentence and absiract level gene gene co citation information was mined by using SciMiner on the complete PubMed abstracts over 21 millions Currently DNMKB performs the sentence level analysis by default Once gene gene co citation networks are generated for the selected DEG sets a graphical analysis tool TALE identifies sub networks shared by two DEG networks Since it takes long time to generate a final graph using TALE TNA will provide users with the URL where the results once ready will be displayed 10 Understanding Result Tables The figure below illustrates an example of the m
14. rted by different criteria such as the number of experiments having each gene as a DEG or by the fold changes in a specific DEG set DEGs from different species will be automatically mapped across different species using the NCBI HomoloGene database http www ncbi nlm nih gov homologene Users should first select datasets to browse Optionally users can adjust the sorting and handling species options before retrieving the results Users can also specify the significance value or minimum fold change cutoff to limit the results to highly significant DEGs If these values are not specified the default values each DEG set have their own default criteria will be automatically used Once all options are chosen users need to click the Generate Matrix button to proceed Diabetic Neuropathy Microarray Knowledge Base Combine multiple species use mouse gene as base Note Combining mu tip e species WOTKS W wo species Generate Matra Resel Significance Cut off lt Minimum Fold Change List of available Microarray Experiments Figure 3 Browse menu Sort by This option specifies how the retrieved DEGs in the result page are ordered The default is dt of experiments with DEGs putting the most frequently perturbed DEGs across multiple conditions on the top list If a specific dataset is selected then the genes will be sorted by the fold change values in the selected dataset The list of sorting options in t
15. t analysis and clustered heat map images of top functions DNMKB provides the analysis results both in text format as well as Excel format facilitating users to perform additional down stream analyses of the DEGs using other tools By default biological functions in terms of GO terms and KEGG pathways with a Benjamini Hochberg BH corrected P value lt 0 05 are deemed significant and will be included in the heat map The heat map will include the top 10 most over represented biological functions in each DEG set clustered based on the significance values log transformed BH corrected P values to visually represent overall 12 similarity and difference between the DEG sets Diabetic Neuropathy Microarray Knowledge Base Functional Enrichment Analysis Functional Enrichment initiated DEG sets Chart full Chart simple Cluster dbub DRG 24wk db vs dbub ChartSumple dxl OID DRU 24wk db vs dbdb Chart sampie ds bdb UNG 24wk db v3 diab Clustin txt dbdb DRG 24wk db v db b fxt KR dbdb DRG 24wk db vs dbdb Chart xis Matrix Text M Matrix Text Iippocampus 24wk db v5 225 The name of DEG set Cluster information lea dbdb Ciuster xis gbdb DRG 24wk db vs dbdb Clusterxis d bdb DRGSwk dgb va dbdb Chari Single txt dixIb DRG Sgwk db ys dbdb Chartsimple o mwk db ys dbdb Cluster txt DRG Swk gb ys dbXb Cluster xis dhabh ORG gwk dte vs
16. tic neuropathy DN related transcriptomics data DNMKB has been developed to facilitate the efficient storage and exploration of the high volume microarray data Table 1 lists the current data sets as of 12 19 2013 including both published and unpublished data While access to unpublished data is currently limited to laboratory members it will also be made publicly available once the associated studies are published Table 1 Overall statistics Total Number Number of experiments Number of DEG sets Statistics DNMKB currently contains 52 differentially expressed gene DEG sets from 13 DN related microarray data sets including transcriptomic profiles in peripheral dorsal root ganglia sciatic nerve and sural nerve and central hippocampus nervous tissues from several mouse models db db BTBR ob ob high fat diet and Streptozotocin induced and human subjects The details of the current transcriptomics data sets are summarized in Table 2 Four microarray data sets from other investigators identified from a public microarray database ArrayExpress http www ebi ac uk arrayexpress were processed by our in house analysis pipeline and included to maximize the data comparability Accessing DNMKB DNMKB is accessible at hitp jdrf neurology med umich edu DNMKB Public users can access any published microarray data while the Feldman Lab members or collaborators have unrestricted access to the database Member login ID and
Download Pdf Manuals
Related Search