Bestand wählen
Merken

Linked Data based Multi-Omics Integration and Visualization for Cancer Decision Networks.

Zitierlink des Filmsegments
Embed Code

Automatisierte Medienanalyse

Beta
Erkannte Entitäten
Sprachtranskript
from the last so far in many ways look and higher knowledge missing that these stick
an inch up of of souls title of my talk is likely that these 2 my arguments integration for instance follow for genomics
these densities in that what's what really think what what what I have done and what I'm doing so I'm going to your soul or leading a political to buy openness that the actor trying to integrate the module makes the Gulf for the American would you put it the new medical discovery so I come from inside center for the for the
analytics and missing they to to miss the University of Island so before getting into the problem explain so if you see the evolution of
clinical decision support system on the medicine in general because the more from evidence visiting the medicine with operation comes not the symmetry phenotypes and that describe somebody from the other of that is the kind of pretty you know it up you know of any evidence these that at the at the 2 that have kind of certain fever so this medicine is required that invent brought up kind of and predicative medicine into the lab last and some some with some reports and so ants and so forth then became public battery-driven it's so in the part FidoNet kind of from all smell from of medicine they went into the error of the discovery their of Dr. going up to the states which is the tentative entity for ducks and the 1 that works T. deal doesn't look now we have the completely genomic and up since 2005 maybe that didn't madam the precision medicine comes into the picture so apart from the the of we need the were all kind of relations if you see the the evaluation in the field of medicine from 1800 if I go back to very back so all of always a medicine have 1st of all a kind of a to be standardized of from the original vision and they went into the kind of before the things that is happening to the very 1st time metagenomic deducting and then there's the achievable opportunity was allows them so there's a lot of actually Aghios in this stage so the so this is the completed a paradigm all for medical easily and we are in this area and this is what we have developed at MIT and we talk about so why do we need
this the the Kurdish inference so we need this for this let us you this is going to be the future officers in medicine it to the paradigm we have a lot of a patient comes to him and he takes the DNA at tightening opera being oral kind of whatever samples he wants and sense to these entities when did this month wonderment condition 1 by the addition and T here we have the sequencing lab which is in the eye of them may be generalized and then they will be up where they will be doing all sort of data on sort of all flow analytics maybe another analytics and then it will reach to the Commission in the formal financial logician and he's going to make it easier to that because we don't get out of all the appointments come from here phosphate fast you fast the formants comes from yet C A 3 at the XT some examined some jocks she he didn't get all that just get about the find out but it should be indirect enough because there's a lot more did in the battle so we need to need to have a clinical decision support most Our comprehensive series addition so here
comes what we have proposed so what we have it was like a a lot more to the of people about them these days of auld with the mix them what what you have is like you have of small or big passing the population you have been with the genomic and they got their genomics and and proteomics condensable midst of of the that you do the or sort of got analytics even we do that we define the single by like and then multiple my mother who didn't look and is made and this and kind of find the light medicine for the light populations this is the eyes where the Onyx we went 1 step before that so there is a very famous concept the this pointer coracoid Concorde italics we extended that the award at though for it's units so in the only slicks more doing let's this data is not enough because anything in in medicine without the analysis of imaging with lots of his eyes so you have the definitively MCandA modality you have your data coming from the from field ecology lab so you have your view again coming from city and if we can coordinate these 2 that you will have a comprehensive solution how the message passing will happen because all because this already done is the neck employment he Haiti you have not adjacent this year's these and Monday performance so here comes the concept of omics what happens in the body it's so be going medical assistance they have be 2 because standards on 7 which is message passing of system from a nurse but doctors through all those steak orders how we extend these in these exchanging messages and then you have been with the performance we have developed something got of only very if we use that in on the day that we recorded on its so this is the most of you have kind of combined at the multiple concepts and blind to bring a comprehensive single point of entry for the medical Domus farther from reality from but this is what we are thinking how does it work so you have
to build an a knowledge of 1st of all in the background how do we know that have so before reading the in the in the in the whole knowledge grafting a sold even that improved devoted to it and a battling what what people have been doing so there have been of following a standard in the horizontal and vertical integration a techniques we went 1 step actually kind of for that let's say we have a single issue of origin for diseases a responsive and some like against both out all the associated kind of cancers so it falls in the top class core disease and cans and they have the same training site now do you have any that any needed entities they might not have an indirect under identities so how do we understand that so let's say of on the the stands and we have a very famous gene called the oven know we have all of gene expression and the that you that the that what we did that here and we revisions of but here we have collected all possible informations and applied to understand the checkpoint then kind of correlates with these these are like a cancer how would it to so once you do your studies will be held it is used and the thing that you do best and so did you to have some correlation in don't sell a to be held the hand the because you be blindly sites is it really alarming nor alarming are they a kind of unity by like that's but we should we ask in in in in in future so we have grown and I love this guy off all of this kind of medical last but they're they're the biggest problem in terms of delays in what we face to the to the complexity of this graph is huge a single gobble 77 do 74 GB so the potluck singled out a system is not going to work for sure so so we should have a distributed of a system so then the year of the Denmark come to us leave and at the end of the day that they are a less money them how do we do that so
before making this kind of all knowledge graph we have to have some rules like with gene is better which proteins that so Foster don't vote for
the example I'm giving like there was a need to be put into a thing into in into the into those and propene and they have select any genomic entity which is being driven by more than 1 genomic achieve concepts is very unlikely to be to be random so that the problems in the medical domain even was that all by what the discovery was only gene expression different now in the early 2 thousand 8 global 2010 gene expression was denied from hot Dathematics and might Lotus which is highly noisy what happens every paper see some of some some some gene were to them as a as a by might we went back and data for the cheapest stands strong decay possibly have 7 for 6 genes if we end up going like this really in the end we will have important too hot all ontologies as as the by much of there that is something pointing there so what would have loved liberty develop some rules or rules so did this for the fostering 2nd room who started
of the cottonwood relief and the intelligence annotation data now we look at the very basic a problems was the what we saw in the 2 by God if like if you have a hyper mutilation you express and this has to be done but there were like more than 200 like the digit more than 2 call coordinates which had the highlights this and high-definition not possible which it's this kind of politically impossible begin foster follow you the idea of making a this information what we have in the graph and this was earlier so we did that so he made such of a
kind of mathematical of Martin's an endorsement of in a 2 year goes with the medical models for these the cheated datasets food to be going to begin with we need develop such rules that apply to Codd knowledge graph and it kind of is the complexity often only got by particles and which was already there indeed the day off from used we then among many the
experiment the project like how and how good how bad our system is so we have this great and the fight and again the blacks very well known that's and it's a very small number so we just how would bother to blacks and in the an indicative of a system which was a part of which we have divided the pool by Foster City might but they have committed a big chunk of data and 2nd was a clock coming from the inside so for every mile you idea for and the that we have of the developed on battle or below a the default if you kind of vinegar then the kid of being forced to do the reality kind of possibly benzidine and at Annex I. the we have developed and then you give it to this this whole system it it it gives you a obtuse costs first-quarter gives you from my if and ejects Lagrangian false in invited 1 and then a text in particle board and then we make it a combined school we have some rules to make a combined score and then the kind of that to standardize the gene led to the politicians and what are the kind of of ML they have at this stage for these but these these 5 doesn't we have 5 a spider and points which given to make life so and so on and so forth but then a 2nd thing was like how solid how how acceptable it would it would 2 would it would be because it has to formal and a formal of some medical model as the so we went back and read
all simple mathematical model that you you have 1 gene a gene then have a transcription and it can have under that and always opportunity you a know for all I know for it can have under and go expression of each with each other for a transcription and read each expression it kind of all at all the king the it kind of it interacts with some importance which the number the more or less the identified through time model with every and the expression to the half becomes so that way we knew that the gene when it comes to Our system how much interaction it should have it has a lot that's a very different picture the cushion but what politicians that we have redefined that so we have we have had this this amendment the medical model it's not most is not the best nectar evoking right it's a very apply many by the lake reelecting noted this very blended it up and I'm not untypical mathematician so what I would do best of my knowledge I have done that it need to be there would be no kind of further validated then the 2nd part was hobbled to the greedy them so here comes the the the
kind of quality so you see it like when you have these or different assists costly could the DCG ICDC BMUG reading of the famous Darfur bloody that which we call up a knowledge graph it to me and the NCAA hazard is no their own official got together and then you have the we all these leaders on these platforms someone would you from I think that among would be be copied the and a budget bill we have now with the new very the yeah the argument was mass that improve it like anything with the new animal picture kind of mapping and the wonderful kind of concept which was formed by Michael Stonebraker as a bully stores which readability Uganda father and make it up on the river then Waterman then in the down your your ordered that is is it to the guy actually kind of going to be with you the Willard Lagueux anything with an OPS despite ready put any kind of data the you the combined results do some some analytics about and in the bag and do this kind of user addition put each type of data so the song of the great since we have already that the kind of love from model for not offered was and impose if the improved the stairs and garbage it for the body of what already
got we have kind of used in this situation at this stage so this is what we have already been kind of using the insulation and this is the Atiyah did this is a typical idea people what we have published in 2016 in the years to be so that these are all the wire and Italian each group can kind of company hand and you comprehends in clinical decision support system so that cannot be a single spike at point that bill bill the and the biggest problem what would be likely to be kind of but I have been knowledge everything it is adjusting the spike and sing this fighting and 1 so we have a lot of time the beauty of our can see is that all of us this time communications and that his view of
the importance of the context initially going to and of the wall of longer them so we have made half idea is feisty is a kind of model by
committee made it because in future with the deconvolution unit what kind of things we just coming I have 1 people that has the red and huge all these in the is as my own lives to put book to kind of clean my model and meet each helium at a school might and then I see but after the the kind of you know after the inclusion of any of these they just how this thing is kind of of changing it is going up because without it is kind of randomly going down so as to have our big a bigger driven by market not to the 2 by market as the on the the expression but then again to this day that we have find this up a bit of 5 billion 6 billion billion tuples then the lake and reuse of Italy all kind of finding out only to finding a proper and that the kind the proper infrastucture who began host them and and that's why we moved up out of the only idea concept and now now we can have jenny date of it and begin to integrate them so this is our opportunity gained 0 or one
hour so promising that all this to review to be on the kind of like the it's really kind of love then all these itchy forced it is and then the hair these kind of it yeah these kind of typical outcomes 2nd problem we face is like ontology and don't there was a lot of it and in the if you such but don't actually kind of chromosomes in by portal relay some holes and see it out just the this the Arts Theatre so we built our own vocabulary that is the of what was known so we have our own knowledge in gain devotedly for dense against against genomics around 55 thousand tons the and then we have kind of been a to the an index of at the moment you send people did that it text on index it is that in the devoted bloody it it it kind of provides differed is not there it will add in that so so this this and then forward with ladies and we need to create
once a month month so any kind of thing so for the gene which coming from 1 source did you know make region that genes involved and so on and so forth Madiba type of leader II obvious than it would if isn't as an actinic and awkward who was a lace so so this is how honored models that you have 1 gene India for it has some attendees information you apply similar and different adding the water quality that's X some stated when you when you have people connections spot on you can extend you got it and underlying years at entity would be a gene or a protein so hands and so what do you exhaust when you lose all the the kind of of of connections than this kind of completes of connection for a gene called media and then which is very important it's selective
of course and once that that which was leave it to the time the hype that the this is the the the kind of thing and that and added what we have been using and
then comes the user addition we had this static image techniques what what we have used was lots of PC and so there is
something called that you have 1 the 1 1 table and you define the 2 that standard of subtypes
that 1 of the kind of for example is this which tended to be the heat so they do this is the guy at the this is the expressions of these other were only enough and you can exons and mindful of what we have in between so it can have the gene you
gender here it can have this kind of genomic to new locations or difference in its prisons yet and different clusters the colligative you know tons of what these what would what what what we can have 2nd thing was know mechanical thing you
at UT and editing and from all those and it got what do you save so we kind of a created orginal make but according to a system for this kind of genomic this kind of genomic data and last but the but not the
least to indicate the imaging of the whole meeting data not plenty of people are only kind of integrating their metadata they're going monster for that we are analyzing these images I'm dying to find the best of Aleksei these 2 kind of what you're reading this of order-dense stop beach and that kind of a to D and biblical bloody age so how the is Devecser is changing the whole concept so so this is it from my got site and I'm really thankful
to kind of inviting for this unit and was envisioned and be offended by signs from his indictment and the other end of collaborating with the the the the the the the how medical school thank you so much
at the few if
anybody wants tools dared if I have time of 1 minute that appeared so we have not by alternate online it is most evident it
had at this stage we are developing it so
if the a let's see if I was I have really if
I Anderson gene here I'm sorry for this let's say and then so these
what happens of art to the students and to the kind of association in order for status and then of its with the state how the gene is kind of correlated with each drug or each corresponding disease and this is a very aluminum dome at this stage is not that standard this is the end of looking at how to make it available so this is what sleep so so you guys can take it's going to be gained by open for the incident of logic and thanks thanks have you all
the questions come to me thank you of mark the
history the questions
Parametersystem
Multiplikation
Linked Data
Datennetz
Desintegration <Mathematik>
Visualisierung
Entscheidungstheorie
Integral
Instantiierung
Multiplikation
Linked Data
Offene Menge
Desintegration <Mathematik>
Datennetz
Analytische Menge
Visualisierung
Modul
Entscheidungstheorie
Dichte <Physik>
Zeitabhängigkeit
Desintegration <Mathematik>
Relativitätstheorie
Entscheidungstheorie
Prädikat <Logik>
Datenfeld
Flächeninhalt
Symmetrie
Mereologie
Protokoll <Datenverarbeitungssystem>
Programmierparadigma
Entscheidungsunterstützungssystem
Maschinelles Sehen
Verkehrsinformation
Fehlermeldung
Aggregatzustand
Leistungsbewertung
Multiplikation
Punkt
Inferenz <Künstliche Intelligenz>
Bildgebendes Verfahren
Desintegration <Mathematik>
Formale Grammatik
Datenmanagement
Analytische Menge
Mathematische Logik
Spezialrechner
Physikalisches System
Einheit <Mathematik>
Elektronische Unterschrift
Datennetz
Stichprobenumfang
Programmierparadigma
Mixed Reality
Notepad-Computer
Zeiger <Informatik>
Chi-Quadrat-Verteilung
Bildgebendes Verfahren
Analysis
Addition
Konkordanz <Mathematik>
Sichtenkonzept
Prozess <Informatik>
Kondensation <Mathematik>
Reihe
Einfache Genauigkeit
Physikalisches System
Datenfluss
Quick-Sort
Einfache Genauigkeit
Office-Paket
Modallogik
Datenfeld
Konditionszahl
Entscheidungsunterstützungssystem
Ordnung <Mathematik>
Message-Passing
Standardabweichung
Beobachtungsstudie
Web Site
Wellenpaket
Quilt <Mathematik>
Graph
Desintegration <Mathematik>
Klasse <Mathematik>
Einfache Genauigkeit
Schlussregel
Physikalisches System
Term
Komplex <Algebra>
Lie-Gruppe
Eins
Axiom
Einheit <Mathematik>
Nichtunterscheidbarkeit
Speicherabzug
Information
Korrelationsfunktion
Standardabweichung
Systemidentifikation
Korrelationsfunktion
Teilmenge
Graph
Betragsfläche
Desintegration <Mathematik>
Datenmodell
Stichprobe
Paarvergleich
Systemaufruf
Regulärer Ausdruck
Vorzeichen <Mathematik>
Schlussregel
Gebäude <Mathematik>
Entscheidungstheorie
Domain-Name
Funktion <Mathematik>
Digitalisierer
Datennetz
Information
Druckertreiber
Ereignishorizont
Retrievalsprache
Korrelationsfunktion
Stellenring
Punkt
Leistungsbewertung
Desintegration <Mathematik>
Zahlenbereich
Statistische Hypothese
Komplex <Algebra>
Datenhaltung
Entscheidungstheorie
Informationsmodellierung
Datennetz
Druckertreiber
Default
Videospiel
Graph
Mathematik
Spider <Programm>
Winkel
Datenmodell
Paarvergleich
Schlussregel
Negative Zahl
Vorzeichen <Mathematik>
Physikalisches System
Schlussregel
Funktion <Mathematik>
Ein-Ausgabe
Mereologie
Projektive Ebene
Partikelsystem
Resultante
Retrievalsprache
Desintegration <Mathematik>
Soundverarbeitung
Zahlenbereich
Interaktives Fernsehen
Systemplattform
Open Source
Arithmetischer Ausdruck
Informationsmodellierung
Mathematische Modellierung
Datentyp
Anwendungsspezifischer Prozessor
Speicher <Informatik>
Addition
Parametersystem
Graph
Transformation <Mathematik>
Datenmodell
Hasard <Digitaltechnik>
Physikalisches System
Lineares Funktional
Automatische Handlungsplanung
Dateiformat
Mapping <Computergraphik>
Rückkopplung
Mathematikerin
Mereologie
Translation <Mathematik>
Lesen <Datenverarbeitung>
Objekt <Kategorie>
Telekommunikation
Punkt
Desintegration <Mathematik>
Gruppenkeim
Einfache Genauigkeit
Kontextbezogenes System
Gesetz <Physik>
Spezialrechner
Informationsmodellierung
OISC
Entscheidungsunterstützungssystem
ICC-Gruppe
Hill-Differentialgleichung
Ereignishorizont
Bit
Multiplikation
Momentenproblem
Ontologie <Wissensverarbeitung>
Konvexe Hülle
Datenanalyse
Desintegration <Mathematik>
n-Tupel
Prognostik
Systemplattform
Oval
Entfaltung <Skalierung>
Einfache Genauigkeit
Arithmetischer Ausdruck
SLAM-Verfahren
Einheit <Mathematik>
Automatische Indexierung
Verkehrsinformation
Ein-Ausgabe
Virtuelle Realität
Hill-Differentialgleichung
Visualisierung
Ontologie <Wissensverarbeitung>
Inklusion <Mathematik>
Einfach zusammenhängender Raum
Multiplikation
Desintegration <Mathematik>
Arithmetische Folge
Anwendungsspezifischer Prozessor
Quellcode
Einfache Genauigkeit
Entscheidungstheorie
Spitze <Mathematik>
Informationsmodellierung
Trennschärfe <Statistik>
Datennetz
Hypermedia
Datentyp
Ein-Ausgabe
Information
Visualisierung
Normalvektor
Tabelle <Informatik>
Systemidentifikation
Distributionstheorie
Addition
Kontrollstruktur
Dichte <Physik>
Desintegration <Mathematik>
Stichprobe
Regulärer Ausdruck
Mathematisierung
Karhunen-Loève-Transformation
Hydrostatik
Arithmetischer Ausdruck
Anwendungsspezifischer Prozessor
Plot <Graphische Darstellung>
Punkt
Visualisierung
Cluster <Datenanalyse>
Bildgebendes Verfahren
Beobachtungsstudie
Normalvektor
Tabelle <Informatik>
Standardabweichung
Gefangenendilemma
Integral
Subtraktion
Geschlecht <Mathematik>
Zellularer Automat
Desintegration <Mathematik>
Anwendungsspezifischer Prozessor
Mapping <Computergraphik>
Physikalisches System
URL
Ontologie <Wissensverarbeitung>
Cluster <Rechnernetz>
Suchmaschine
Determinante
Metadaten
Web Site
Motiv <Mathematik>
Einheit <Mathematik>
Verbandstheorie
Bildgebendes Verfahren
Vorzeichen <Mathematik>
Desintegration <Mathematik>
Zeiger <Informatik>
Bildgebendes Verfahren
Desintegration <Mathematik>
Retrievalsprache
Desintegration <Mathematik>
Speicherabzug
Gauss <Rechenmaschine>
Lineares Funktional
Assoziativgesetz
Retrievalsprache
t-Test
Elektronischer Datenaustausch
Ordnung <Mathematik>
Inzidenzalgebra
Mathematische Logik
Korrelationsfunktion
Standardabweichung
Aggregatzustand
Desintegration <Mathematik>

Metadaten

Formale Metadaten

Titel Linked Data based Multi-Omics Integration and Visualization for Cancer Decision Networks.
Serientitel Data Integration in the Life Sciences (DILS2018)
Autor Jha, Alokkumar
Khan, Yasar
Mehmood, Qaiser
Rebholz-Schuhmann, Dietrich
Sahay, Ratnesh
Lizenz CC-Namensnennung 3.0 Deutschland:
Sie dürfen das Werk bzw. den Inhalt zu jedem legalen Zweck nutzen, verändern und in unveränderter oder veränderter Form vervielfältigen, verbreiten und öffentlich zugänglich machen, sofern Sie den Namen des Autors/Rechteinhabers in der von ihm festgelegten Weise nennen.
DOI 10.5446/38890
Herausgeber Technische Informationsbibliothek (TIB)
Erscheinungsjahr 2018
Sprache Englisch

Inhaltliche Metadaten

Fachgebiet Biowissenschaften / Biologie, Informatik
Abstract Visualization of Gene Expression (GE) is a challenging task since the number of genes and their associations are diffcult to predict in various set of biological studies. GE could be used to understand tissuegene- protein relationships. Currently, Heatmaps is the standard visualization technique to depict GE data. However, Heatmaps only covers the cluster of highly dense regions. It does not provide the Interaction, Functional Annotation and pooled understanding from higher to lower expression. In the present paper, we propose a graph-based technique - based on color encoding from higher to lower expression map, along with the functional annotation. This visualization technique is highly interactive (HeatMaps are mainly static maps). The visualization system here explains the association between overlapping genes with and without tissues types. Traditional visualization techniques (viz-Heatmaps) generally explain each of the association in distinct maps. For example, overlapping genes and their interactions, based on co-expression and expression cut off are three distinct Heatmaps. We demonstrate the usability using ortholog study of GE and visualize GE using GExpressionMap. We further compare and benchmark our approach with the existing visualization techniques. It also reduces the task to cluster the expressed gene networks further to understand the over/under expression. Further, it provides the interaction based on co-expression network which itself creates co-expression clusters. GExpressionMap provides a unique graphbased visualization for GE data with their functional annotation and associated interaction among the DEGs (Differentially Expressed Genes).

Zugehöriges Material

Ähnliche Filme

Loading...
Feedback