Taylor: All CTNNB1 and BRAF mutations are clonal by their analyses; the only putative clonal drivers #AACR14
6:53pm April 6th 2014 via Hootsuite
Taylor: Found the two subtypes characterized by CTNNB1 driver or BRAF V600E driver. Mutually exclusive between types #AACR14
6:52pm April 6th 2014 via Hootsuite
Taylor: 15 exome samples, 89 add'l for targeted; due to rarity taken from many locations. TCGA chart from http://t.co/sEvmJLaHHf #AACR14
6:51pm April 6th 2014 via Hootsuite
Taylor: Craniopharyngiomas are rare (0.18 per 100K), no effective therapy, complications common due to location (neurological fns) #AACR14
6:49pm April 6th 2014 via Hootsuite
Next: Amaro Taylor-Weiner, Broad Inst. "Exome sequencing reveals BRAF mutations in papillary craniopharyngiomas" #AACR14
6:48pm April 6th 2014 via Hootsuite
.@milzpg You are most welcome!
6:45pm April 6th 2014 via Hootsuite in reply to
RT @AACR: See the #EmperorOfAllMaladies preview we played at the #AACR14 opening ceremony: http://t.co/W4ZRLn4rTa @SU2C @KenBurns
6:40pm April 6th 2014 via Hootsuite
Boussemart: 66 samples, severa; were extra-cutaneous; pERK all +; BRAF/CRAF dimerization by proximity ligation assay #AACR14
6:37pm April 6th 2014 via Hootsuite
Boussemart: BRAF inh. treatment cause skin secondary effects; 20% squamous cell carcinomas, secondary melanomas #AACR14
6:35pm April 6th 2014 via Hootsuite
Next: Lise Boussemart, Roussy: BRAF inhibitors induce skin and extra-cutaneous tumors via paradoxical activation of the MAPK pathway #AACR14
6:33pm April 6th 2014 via Hootsuite
Wilson: Hypothesis: sorafenib worked through a cRAF inhibitor. RAF1 amplification thus improves responsiveness #AACR14
6:28pm April 6th 2014 via Hootsuite
Wilson: Looking at 25 genes with copy gains / losses, looking at BRAF amplification 77% of samples (aCGH); ass'd with somatic mut's #AACR14
6:26pm April 6th 2014 via Hootsuite
Wilson: Looking at melanoma, ECOG2603; reference Flaherty et al JCO 2013 Pubmed: http://t.co/vJYSXuD6FB #AACR14
6:23pm April 6th 2014 via Hootsuite
Next: M Wilson U Penn "CN changes are assoc'd with BRAF and NRAS mut's and resp to treatment with carboplatin, paclitaxel and sorafenib"
6:20pm April 6th 2014 via Hootsuite
Boutros:30 phenotypes measured with 1200 bioactive compounds, looking at chemical - genotype interactions #AACR14
5:57pm April 6th 2014 via Hootsuite
Boutros: Took isogenic cell lines, screened against thousands of compds, used image-based phenotyping, 240K datapoints #AACR14
5:56pm April 6th 2014 via Hootsuite
RT @AppliedBio: Let's have that conversation about laser capture microdissection. #AACR14 Booth 1415 http://t.co/g3aKrCeCh2
5:54pm April 6th 2014 via Hootsuite
Boutros:Can go to a 'wiring diagrams', also score on co-complexity to predict protein machinery / complexes #AACR14
5:52pm April 6th 2014 via Hootsuite
Boutros:Plot of distances of genetic interaction profile, based on similarity of genetic interaction profiles. #AACR14
5:51pm April 6th 2014 via Hootsuite
Boutros: Use 2 RNAi/gene, measure all effects. Do single-cell phenotyping (visual), geometry, many dimensions #AACR14
5:50pm April 6th 2014 via Hootsuite
Boutros:They use gene-gene interactions - quantitative double mutant analysis. 1.3k genes, double perturbations + phenotypes (500K) #AACR14
5:49pm April 6th 2014 via Hootsuite
Boutros: The 'hidden activity of modules, pathways, complexes', reminiscent of Califano's 'black box' between g-type and p-type #AACR14
5:44pm April 6th 2014 via Hootsuite
Next: Michael Boutros, German Cancer Res Ctr Heidelberg "Mapping genetic interactions in cancer pathways" #AACR14
5:42pm April 6th 2014 via Hootsuite
This morning's 'Emporer of All Maladies' video now on the #AACR14 YouTube channel http://t.co/USq5SHvT4I (I'm going to send this to friends)
5:39pm April 6th 2014 via Hootsuite
Ideker:Q:Doesn't exome data as input make data more noisier? A:Mut's are a binary call cp to exp. data, will integrate exp later #AACR14
5:37pm April 6th 2014 via Hootsuite
Ideker: You can't predict structure from network data. Quoted from this Nature Methods (2013) PubMed: http://t.co/8NCrLeW8Ru #AACR14
5:34pm April 6th 2014 via Hootsuite
Ideker: Where to go from here? Networks are flat; biological structures are not. Illus. a proteosome vs. its network diagram #AACR14
5:31pm April 6th 2014 via Hootsuite
RT @jennamanenna: Teenage cancer survivor/researcher Elana Simon receives @AACR Jr Champion Cancer Res Awrd #AACR14 http://t.co/yLK8bjMva7””
5:27pm April 6th 2014 via Hootsuite
Ideker: In ovarian ca - plotting survival data; derived 4 clusters, showed clear differences in survival plots #AACR14
5:24pm April 6th 2014 via Hootsuite
Ideker: Network smoothing paper in PLoS Computational Biol PubMed http://t.co/8kmPONDTU2 #AACR14
5:22pm April 6th 2014 via Hootsuite
Ideker: Uses bootstrapping clustering framework & network smoothing: sparse signals now meas. by distance in pathway space #AACR14
5:21pm April 6th 2014 via Hootsuite
Ideker: Network stratification tools: Pathway commons (MSKCC, Sander), HumanNed (UT Austin, Marcotte), StringDB (EMBL, Bork) #AACR14
5:19pm April 6th 2014 via Hootsuite
Ideker: Clustering rare events across a population when they are so different? A disease of pathways. A subway map #AACR14
5:18pm April 6th 2014 via Hootsuite
Ideker: Other work on mut calling & driver pred. - they wrok on stratificaiton when no two tumor exomes are the same #AACR14
5:15pm April 6th 2014 via Hootsuite
Ideker:Network using exome data: how are mut's called? (80-100 sample); what drivers are predicted?; stratification - how? #AACR14
5:14pm April 6th 2014 via Hootsuite
Ideker: Systems biology between g-type/p-type - will be talking about the construction of 'Cytoscape' network http://t.co/pGI4X3GKoq #AACR14
5:13pm April 6th 2014 via Hootsuite
Up next: Trey Ideker, UCSD "Network based stratification of tumor mutations". #AACR14
5:11pm April 6th 2014 via Hootsuite
Califano:Q:Looked at indiv differences? A:Every sample treated independently, all 'N of 1' protocol. But networks only change ¬5%. #AACR14
5:09pm April 6th 2014 via Hootsuite
Califano:Q:Do MRs populate super enhancers? A:Disregulation of myc or others can occur in many ways. #AACR14
5:06pm April 6th 2014 via Hootsuite
Califano: Alterations of genes and affect MRs? Using EGFR as example, can associate spec mutations to master regulators #AACR14
5:04pm April 6th 2014 via Hootsuite
Califano: Across 5.4K TCGA samples, can look at master regulators across subtypes. #AACR14
5:01pm April 6th 2014 via Hootsuite
Califano: Looking now at mutations that may affect master regulators.10K var, 3.5k samples, 12 tumor types, 5K regulators each ass'd #AACR14
4:57pm April 6th 2014 via Hootsuite
Califano: Looking at Foxm1, Cenpf as master regulators, combination of observed compounds, model shown to work as expected #AACR14
4:54pm April 6th 2014 via Hootsuite
Califano: Prostate ca, 13 transgenic mice, hit with 28 compounds, 364 matrix. Of top 20 regulators, 6 from mouse intersect w/ human #AACR14
4:49pm April 6th 2014 via Hootsuite
Califano: A phenotype of interest - what genes regulated the genes that changed? So they can rank the regulators #AACR14
4:45pm April 6th 2014 via Hootsuite
Califano: Types of big data - transcr. interactions, post-transl. interactoins, post-transcriptional interactions, master regulators #AACR14
4:44pm April 6th 2014 via Hootsuite
Califano: Looking at 3-5 regulatory proteins that cause a 'cancer bottleneck' - oncogene-induced dependencies #AACR14
4:42pm April 6th 2014 via Hootsuite
Califano: 'Black Box' (colored in green!) - open it up and look at the 'ridiculomes' with cellular context of regulatory models #AACR14
4:40pm April 6th 2014 via Hootsuite
Califano: But events are complex: Trastuzumab 70% response, then 70% relapse in br ca #AACR14
4:39pm April 6th 2014 via Hootsuite
Califano: Linking a gene or var to phenotypic outcome. SOD1 in ALS (not targetable), BRC/Abl fusion resulted in Imatinib, Nolotinib #AACR14
4:38pm April 6th 2014 via Hootsuite
