Integrative precision medicine: Local Data, Global Context Sven - PowerPoint PPT Presentation


 
 Integrative precision medicine: Local Data, Global Context Sven Nelander Associate Professor Dept of Immunology, Genetics and Pathology 
 Uppsala University

Two fundamental challenges for cancer research Drug X! Drug X? Drug X?? Local cohort / patient data ( n <1000) U-CAN HGCC UAS BILS TCGA L1000 SUS SCAN-B LINCS ICGC KS 100k genomes Global background data (n>100,000)

Examples of global data: The Cancer Genome Atlas and LINCS/L1000 0 500 transcripts 1000 1500 0 1000 2000 3000 4000 5000 6000 7000 drug

MG−132 U3065_95 U3179 U3291 U3279 U3289 U3299 U3009 U3029 U3129 U3039 U3019 Kruskal−Wallis p: 0.49 man: 0.079 p: 0.57986 W GAP_JUNCTION 1e−05 YS_IN_CANCER 4e−05 Example of local data: differential drug responses in Swedish brain tumors 4.2e−11 4.2e−11 6.1e−09 5.2e−07 Mean viabiliy (Carfilzomib) < 2.22e−16 1.1e−05 Mean viabiliy (MG-132) 1.5 1.5 < 2.22e−16 3.4e−05 3.6e−05 1.0 1.0 Carfilzomib MG-132 0.5 0.5 D 0.0 0.0 0.001 0.01 0.1 1 10 100 i ii 0.001 0.01 0.1 1 10 100 Signature drugs Dose (uM) Dose (uM) Drug Cluster U3179 PSMB5 Bortezomib Oprozomib TOP2B Mean viabiliy (Oprozomib) U3179 TUBB Targets TUBB4B Mean viabiliy (Ixazomib) 1.5 TUBA1A 1.5 DRD2 HTR2A U3179 U3179 sex Associations subtype U3179 1.0 1.0 MGMTmeth U3179 0.5 0.5 0.0 0.0 0.001 0.01 0.1 1 10 100 0.001 0.01 0.1 1 10 100 Dose (uM) Dose (uM) Cell lines Subtype Cluster 0 1 Drugs AUC score

Goals and team 1. To develop computational methods for integrative modeling of big cancer data sets 2. To integrate small-cohort data and global modeling for target discovery in cells from patients 3. To make computational tools available as a web portal and efficient standalone software Terry Speed Sven Nelander Rebecka Jörnsten Björn Nilsson Erik Sonnhammer George Michailidis Biostatistics Cancer systems biology Mathematical statistics Systems Medicine Bioinformatics Machine Learning Walter and Eliza Hall Institute 
 Uppsala University Chalmers Technical U Lund University Stockholm University University of Florida UC Berkeley

Project organisation Deep models (1.1) Rebecka Jörnsten (co-lead PI) George Michailidis (collaborator) Jonathan Kallus (PhD student) Postdoc Szilard Nemes (postdoc) ++ New PhD student ++ Shared Resources Data integration (GitHub, server, AWS) methods (1.1,1.2) Erik Sonnhammer (co-PI) Sven Nelander (lead PI) Analysis of Daniel Morgan (PhD student) Patrik Johansson (PhD student) response Deniz Secilmis (PhD student) ++ Emil Rosén (PhD student) ++ data (2.1) New postdoc ++ Caroline Wärn (bioinformatician) ++ Anders Sundström (bioinformatician) ++ Cecilia Krona (researcher) Terry Speed (collaborator) Elin Almstedt (PhD student) Ingrid Lönnstedt (reseacher) New Postdoc ++ Cell response Shared Resources RNA-seq (2.2) (molecular biology) Björn Nilsson (co-PI) Color Key: Ludvig Ekdal (PhD student) Computational biologist Mathematician Maroulio Pertesi (researcher) Experimental systems biologist Ram Ajore (researcher) ++ = recruited as part of SSF program implementation

Aim 1: new data integration methods A Individual TCGA Uppsala patient cohort cohort shared co-variation across cohorts data modalities shared co-variation across data modalities Jörnsten et al, MSB 2011; Kling et al, NAR 2015; Kling et al EBioMedicine 2016; 
 Schmidt et al, OncoTarget 2016; Yiang et al, Cell Reports 2017

Aim 2: demonstration study on brain tumor cells from Swedish patients Drug response profiles 
 L1000/LINCS (extreme scale profiling of (collected at UU,LU drugs in 77 cell lines) by our team) Calibration study 0 0 500 500 transcripts transcripts Machine learning models 1000 1000 Use of models to 1500 1500 predict interventions 0 1000 2000 3000 4000 5000 6000 7000 0 1000 2000 3000 4000 5000 6000 7000 drug drug

Aim 3: new tools for cancer data mining Interactive motif drawing MATCH (n1)-[e00:HGCC]-(n0), (n0)-[e2]-(n3), (n3)-[e4]-(n5) WHERE n0:isTF AND n0.type="expr" AND n1.type="drug" AND n3.type="cna" AND n5.type="clinical" WITH (e00.weight+e2.weight+e4.weight) as score, n0, n1, n3, n5, e00, e2, e4 RETURN score, n0, n1, n3, n5, e00, e2, e4 ORDER BY score DESC Ranking and relation of matching motifs 1 Patrik Johansson 2 3 Johansson et al, in preparation

Current status and plans onwards 3 workshops held + 1 international meeting hosted (SCAS) • 1 project paper accepted, 2 in preparation • Research visits Uppsala <-> Gothenburg and Lund • Discussions with SciLifeLab leadership on next-generation data integration • Total team grown by 6 new members (UU, SU, CTH). 2 recruitments to go • Hazard Hazard MYCN MYCN Stage Stage ALK ALK 1 aggregate scoring across cohorts and cell models 1 AKT1 hits EGFR FDR=5% ERBB2 FDR=10% MAPK8 MTOR p=10 -20 FDR=20% TP53 ALK MAPK1 Stage MAPK3 chemical compound score CASP8 0 HIF1A JUN TUBB4B BCL2 target CASP2 CASP7 500 analysis CDK2 transcripts CFLAR 2 FDR=30% GSTP1 p=10 -10 HDAC1 Hazard PARP1 VEGFA 1000 CASP3 ABCB1 ABCG2 CYP3A4 ERBB3 IGF1R 1500 JAK2 MCL1 0 1000 2000 3000 4000 5000 6000 7000 MMP9 MYCN drug 0 SRC 3 1. RNA and clinical data 2. RNA response profiles 3. Aggregated scores of 4. Enriched protein from multiple cohorts of >7000 drugs in 77 cell lines small molecules targets Big data processing pipeline Lönnstedt et al, SAGMB 2017 Almstedt et al, in preparation Johansson et al, in preparation