Ideker: Analysis challenge: 'networks do not look like contents of cells' and shows diagram of proteasome, and an X-ray structure #AACR16
8:28am April 20th 2016 via Hootsuite
Ideker: Started another effort called Data4cure https://t.co/iRLUgldtrP #AACR16
8:27am April 20th 2016 via Hootsuite
Ideker: Now aggregating this kind of network data '15 ref https://t.co/uy80e4teii website https://t.co/mEzmYafBpK #AACR16
8:26am April 20th 2016 via Hootsuite
Ideker: Network aggregating low-survival ov cancer pts in TCGA; shows comparison of two individuals by network nodes #AACR16
8:25am April 20th 2016 via Hootsuite
Ideker: More recent work - ICGC stratification, 3y OS similar across subtypes. #AACR16
8:23am April 20th 2016 via Hootsuite
Ideker: Shows data for drug resistance by subtype, Platinum-based Tx; subtype 1 is almost all resistant ('13 Nat Methods ref) #AACR16
Ideker: FGFR receptor family, red cluster linked closely; shows OS curves by subtype ID'd by stratification #AACR16
8:21am April 20th 2016 via Hootsuite
Ideker: Robust clusters due to raw data intersected with network knowledge; 'heat' upstream or downstream affects rel's #AACR16
8:20am April 20th 2016 via Hootsuite
Ideker: Can get a nice network-level. Approach borrowed by other science domains: network smoothing https://t.co/mW0SQUFqfV #AACR16
8:18am April 20th 2016 via Hootsuite
Ideker: If heterogeneous genomes are integrated at a higher-order, at the network level (30K proteins, 0.5M links between them) #AACR16
8:17am April 20th 2016 via Hootsuite
Ideker: Network-based stratification of genomes '13 Nat Meth https://t.co/koFwsAyV1I sources: Pathway Commons, HumanNet, StringDB #AACR16
8:16am April 20th 2016 via Hootsuite
Ideker: Thus tumor supp genes and druggable targets. ING5 chr remodeling gene, then target TUBA1A (cytoskeleton) where Rx exists #AACR16
8:13am April 20th 2016 via Hootsuite
Ideker: Tackling two genes at once - a suppressor gene knocked out, then therapy strategy is to find a second gene to knock down #AACR16
8:12am April 20th 2016 via Hootsuite
Ideker: 2nd approach is genetic interaction; the synthetic lethal interaction. Logical relationships between pairs of genes. #AACR16
8:11am April 20th 2016 via Hootsuite
Ideker: MS to ID the unk interactors, build network of a hub with interacting proteins ID'd #AACR16
8:10am April 20th 2016 via Hootsuite
Ideker: Using AP/MS/MS: express tagged protein (FLAG or other), expressed in cell line (or panel of lines); pull-down partners #AACR16
Ideker: Cancer Cell Map initiative https://t.co/BbJKaDxtHi and look at how networks can be mapped for discovery, Rx #AACR16
8:09am April 20th 2016 via Hootsuite
Ideker: By discovery type - gene manipulation by Knock Down (KD), knock in etc the highest yield for discovery (some 267 genes) #AACR16
8:08am April 20th 2016 via Hootsuite
Ideker: Genome analysis did not originally discover most known cancer genes. Shows chart of first publ; 90% already known #AACR16
8:07am April 20th 2016 via Hootsuite
Ideker: AKT, MAPK, both involve RAS even though only 1% in GBM. But how to integrate pathway info systematically? #AACR16
8:05am April 20th 2016 via Hootsuite
Ideker: Vertical bar w/TP53; but everything else is completely non-recurrent. Integrate events into networks https://t.co/l5XSr0iFTe #AACR16
8:04am April 20th 2016 via Hootsuite
Ideker: There are more than 10K cancer WGS; but heterogeneity shows itself. TCGA ov ca n=351 shown for Ch17 by 550 genes #AACR16
8:03am April 20th 2016 via Hootsuite
Trey Ideker (Univ CA San Diego, CA) The Cancer Cell Map Initiative #AACR16
8:01am April 20th 2016 via Hootsuite
In a prior role at Thermo, I wrote up Trey Ideker now at #AACR16 here. https://t.co/hXXgj1Km6G Meet the expert in Rm 260, 7am Wed
7:59am April 20th 2016 via Hootsuite
.@GenomeBiology @LaoSaal Indeed - thanks for the reference!
5:49am April 20th 2016 via Hootsuite in reply to GenomeBiology
MT @Nikhilwagle: Using social media to partner w patients to study metastatic breast cancer https://t.co/jP1OpKntIb #mbcproject #aacr16
5:47am April 20th 2016 via Hootsuite
RT GuneetWalia: Regev: fantastic data on Single Cell RNA-Seq fr #PrecisionMedicine underway fr 21 cancers #AACR16 https://t.co/VTz6kGSuWG
5:42am April 20th 2016 via Hootsuite
RT @broadinstitute: At #AACR16? Aviv Regev and Levi Garraway's single-cell approach to look inside tumors: Video: https://t.co/avd9pAwtJG
5:35am April 20th 2016 via Hootsuite
RT @thakkars: Barb Conley discussing NCI-MATCH interim results @theNCI @NCItreatment #AACR16 https://t.co/nttjat0bec
5:32am April 20th 2016 via Hootsuite
@LaoSaal (i.e. click 'view abstracts' from there, and then do an advanced search for Mortimer or Abstract 506.)
5:23am April 20th 2016 via Hootsuite in reply to LaoSaal
.@LaoSaal Mortimer S et al is currently under review. Can't deep link to it - see ihttp://ow.ly/4mTcRG and need to search for it.
5:22am April 20th 2016 via Hootsuite in reply to LaoSaal
.@LaoSaal Alas Lao, while there were about 8 people sitting in the 5 rows in front of me taking snapshots of the slides, I wasn't able to.
5:20am April 20th 2016 via Hootsuite in reply to LaoSaal
RT @AACR: We welcome @VP Biden to #AACR16! 12:15pm Wed, April 20 Watch the live webcast: https://t.co/meQzAmKpqH https://t.co/LLAsTYGoA8
11:14pm April 19th 2016 via Hootsuite
The list of cancers that can be treated by immunotherapy keeps growing - The Washington Post https://t.co/nx6bsvvX8i #AACR16 news
11:04pm April 19th 2016 via Hootsuite
Q: Normal people? Zill: Need to be carefully considered, but yes (in the future) #AACR16
5:43pm April 19th 2016 via Hootsuite
Q: Can you go down to single copy mutant molecule? Zill: 'short answer is yes' #AACR16
5:42pm April 19th 2016 via Hootsuite
Q:44% of Stg IV? Zill: 44% in 8 genes, due to mixed cohort. Stg III? Don't really know, no more info provided. #AACR16
5:41pm April 19th 2016 via Hootsuite
Zill: Pts w/sensitizing mutations in plasma ctDNA at <0.1% mutant have shown response to Rx #AACR16
5:39pm April 19th 2016 via Hootsuite
Zill: A global summary in metastatic colorectal cancer, and wide array of mutations and their frequencies shown (adv stage) #AACR16
Zill: ALK intron 19, EML4 intron 9, 16, 20 freq shown. Con't w/indiv pt with 0.06% EML4-ALK fusion in ctDNA, nearly complete resp #AACR16
5:38pm April 19th 2016 via Hootsuite
Zill: Ampl and fusion patterns in ctDNA 'mirror tissue'; Breast, colorectal shown with high correlation. Fusions ALK/EML4 too #AACR16
5:36pm April 19th 2016 via Hootsuite
Zill: KRAS and PIK3CA, both r=0.99. EGFR (NSCLC only) r=0.76, r=0.91 excluding T790M (due to pretreatment) #AACR16
5:34pm April 19th 2016 via Hootsuite
Zill: r=0.94 across 4k var's in TP53 between Guarant and TCGA. #AACR16
Zill: ctDNA cohort - adv stage, previously-treated later stages for Guardant, but TCGA has all stages I-IV. Look at TP53 first #AACR16
5:33pm April 19th 2016 via Hootsuite
Zill: Used cBioPortal TCGA data, filter their data to 70 genes, looked at freq, mutatio corelation, mutual exclusivity analysis #AACR16
5:31pm April 19th 2016 via Hootsuite
Zill: Detection rate across liver, CUP, bladder etc - n=5,240, ave 85% detection rate. GBM only 57% due to blood-brain barrier #AACR16
Zill: Half of reported var's 'occur below 0.4% MAF'. Chart of var freq - 25th percentile was 0.18%, median 0.39% #AACR16
5:30pm April 19th 2016 via Hootsuite
Zill: Seq depth, tagging eff, somatic calling, indels, CNVs, fusions. Showed tumor response map. #AACR16
5:29pm April 19th 2016 via Hootsuite
Zill: 150kb panel, ave 15K-x depth (read). 0.06% MAF 3/4800 copies. Mortimer et al (in review) Combines tagging, statistical filter #AACR16
5:28pm April 19th 2016 via Hootsuite
Zill: But to-date, population-scale unk. Platform: critical exons in 70gnes, 5-30ng, 1500 - 9990x #AACR16
5:27pm April 19th 2016 via Hootsuite
