Deihn: but w/ 8 instead of 1 mutation; pushe down to <0.01%. Overlaying error rate, background error is significantly high #TRICON
11:56am March 10th 2016 via Hootsuite
Deihn: Probability model based on data. Newman et al Nature Biotechnol (in press) curve shown: 90% probability = 0.02% #TRICON
11:55am March 10th 2016 via Hootsuite
Deihn: Biggest impediment to improving LoD: error rate of DNA sequencing, including PCR step #TRICON
11:54am March 10th 2016 via Hootsuite
Deihn: How to improve LoD: # of molecules recovered (more blood - from 1.5mL now 5mL plasma). # mutations: modify tergeted pael #TRICON
11:53am March 10th 2016 via Hootsuite
Deihn: ctDNA analysis correlated with imaging: pg/mL ctDNA vs tumor volume at 4 mos and 18 mos. Surgery example also #TRICON
11:50am March 10th 2016 via Hootsuite
Deihn: Shows monitoring pt w/Stg IV NSCLC, over course of 16 months assoc'd with therapy #TRICON
11:49am March 10th 2016 via Hootsuite
Deihn: At the lowest level, less than 1/3 of single cell's worth of DNA (6pg). Assoc between ctDNA conc and tumor volume: r2 = 0.89 #TRICON
11:48am March 10th 2016 via Hootsuite
Deihn: Shows ROC curve, ctDNA bet 0.019% - 3.2%. ctDNA con of 1.9 - 226 pg/mL in Stage I-IV NSCLC #TRICON
11:47am March 10th 2016 via Hootsuite
Deihn:Can monitor; '14 Nature Med ref: https://t.co/pSfz335XC2 Workflow uses Roche NimbleGen SeqCap; 2.5K - 10K-fold coverage #TRICON
11:46am March 10th 2016 via Hootsuite
Deihn: Cancer Personalized Profiling by Deep Seq (CAPP-Seq) - targets recurrently mutated regions; high depth (det 0.01%), SVs, CNVs #TRICON
11:45am March 10th 2016 via Hootsuite
Deihn: dPCR can be ddPCR, BEAMing. WES/WGS to cfDNA detection is not that good. #TRICON
11:43am March 10th 2016 via Hootsuite
Deihn: (Meant to write ASP - allele-specific PCR); WES 5%; WGS 1%; allele-spec PCR at 0.1-0.05% dPCR 0.01% #TRICON
11:42am March 10th 2016 via Hootsuite
Deihn: Need broad pt coverage - unknown which mutations before testing. Graph detection limits - Sanger at >10%, Pyro ASE 10% #TRICON
11:41am March 10th 2016 via Hootsuite
Deihn: 1pg - 10ng/mL range: need sensitive assays. Limited input; Stg IV is 1-2% of cfDNA; need to capture high % molecules #TRICON
Deihn: Ranges for plasma constituents - Alb Hgb highest, 1-10ng/mL is PSA, CEA. Below is Estradial, thyroid hormone FT3. ctDNA spans #TRICON
11:39am March 10th 2016 via Hootsuite
Deihn: cfDNA has a half-life of 0.5-2h; shows size distribution w/small bump at ~350bp, another small one ~525bp #TRICON
11:38am March 10th 2016 via Hootsuite
Deihn: Properties: major is ~175bp; typically ~5ng/mL plasma in healthy adults; higher can be seen in cancer, also infection, trauma #TRICON
11:37am March 10th 2016 via Hootsuite
Deihn: Potential: non-invasive genotyping, monitor trtmnt; ID MRD; surveillance; ID resistance muts; early detection #TRICON
11:36am March 10th 2016 via Hootsuite
Deihn: ctDNA is attractive as a biomarker b/c: mut specific to tumor cells; involved in pathogenesis; some are actionable #TRICON
11:35am March 10th 2016 via Hootsuite
Max Deihn (Stanford) Deep sequencing of circulating tumor DNA for personalized cancer detection and monitoring #TRICON
Garlick: Run, cluster density, number of reads, Q3 scores, frequency. #TRICON
6:55pm March 9th 2016 via Hootsuite
Garlick: Working on an Integrated Quality software: can estab. QC metrics with uploaded VCF files; customizable; can troubleshoot #TRICON
6:54pm March 9th 2016 via Hootsuite
Garlick: Outliers outside 95% binomial prediction: can see difference of individual labs. Range 4.4% to 21.2% #TRICON
6:51pm March 9th 2016 via Hootsuite
Garlick:Inter-lab, same platform: able to make statistically-relevant comparisons. Average target freq/measuring distribution #TRICON
6:50pm March 9th 2016 via Hootsuite
Garlick: Results from multi-lab study using 10% and 20% allele frequency, w/26 mutations shows box-plot and 2% higher on Ion Torrent #TRICON
6:48pm March 9th 2016 via Hootsuite
Garlick: SeraCare offers solid tumor mutation mix at a variety of allele frequencies, has cell-free tumor DNA, designed to challenge #TRICON
6:44pm March 9th 2016 via Hootsuite
Garlick: Highlights from the FDA NGS Oncology Panels workshop - pt sample is the internal DNA ctl; need for high multiplexed ctls #TRICON
6:43pm March 9th 2016 via Hootsuite
Garlick: Challenges: tech advancing faster than the QC; lack of assay/platform standardization; implications of incorrect results #TRICON
6:39pm March 9th 2016 via Hootsuite
Garlick: Starts with the statement: "Precision Medicine is about the Patient" #TRICON
6:37pm March 9th 2016 via Hootsuite
Russell Garlick (SeraCare) NGS-based diagnostics: developing assays and monitoring performance using novel biosynthetic QC tools #TRICON
6:36pm March 9th 2016 via Hootsuite
Daber:"It would be great if the community could come up with some answers (to this problem)" #TRICON
6:31pm March 9th 2016 via Hootsuite
Daber: Could be technical artifact (high expressor depleted assay). So they cp to FISH, but can't get hard quant numbers. #TRICON
Daber: Sens challenges:TN vs FN. No fusion reads, no WT reads: could be biological, could be technical. #TRICON
6:30pm March 9th 2016 via Hootsuite
Daber: WT expression, dope in synthetic xcrpt, capture correlates w/input. Fusion counts a few at 16/cell. 'Read capture correlates' #TRICON
6:29pm March 9th 2016 via Hootsuite
Daber: LOD: do mixed chimeric samples work? DNA works when loci have equal copy number. Variant freq DNA okay; RNA? #TRICON
6:28pm March 9th 2016 via Hootsuite
Daber: Obtaining controls: 'one of the biggest challenges to validating fusion panels is obtaining materials' #TRICON
6:27pm March 9th 2016 via Hootsuite
Daber: Can use RNA/DNA 'to lock down the biology'. Genome equivalents to RNA copies. Accuracy cp to FISH; precision, sens/reprod #TRICON
Daber: Solid tumor panel has 50 markers, Lung/Thyroid has 6. Total NA: both RNA and DNA reads, if fusion not-present will see that #TRICON
6:26pm March 9th 2016 via Hootsuite
Daber: Archer tech: needed two rounds w/nested primer gave higher on-target. Shows importance of unique molecules #TRICON
6:24pm March 9th 2016 via Hootsuite
Daber: Archer assay is semi-amplicon: one random, one known start. Hybrid of traditional library prep, and amplicon-spec targeted #TRICON
6:23pm March 9th 2016 via Hootsuite
Daber: QC after RT using Archer assay: using Archer preSeq assay. #TRICON
6:22pm March 9th 2016 via Hootsuite
Daber: Shows Bioanalyzer scans, and how they segregate peak areas, integrate, and have data supporting quality of certain areas #TRICON
6:20pm March 9th 2016 via Hootsuite
Daber: Yield of DNA/RNA may be variable. QC via Tapestation, Qubit, qPCR: FFPE DNA 'is not equal in quality' #TRICON
Daber: Took benign tissue, combined with mouthwash and compared. Their SOP: deparaffin w/QIAGEN Depara sol'n.2nd hi-temp water elute #TRICON
6:18pm March 9th 2016 via Hootsuite
Daber: Total NA advantage: conserve tissue, test both RNA/DNA. Cons: decrease total yield cp to each separately #TRICON
6:16pm March 9th 2016 via Hootsuite
Daber: Purified RNA vs total Nucleic Acid? Or use both to help quantitate RNA to DNA ratios?(Normalize to genomic equivalents?) #TRICON
6:14pm March 9th 2016 via Hootsuite
Daber: DNA-based rearrangements can be difficult - so many forms. Illus from '13 ref https://t.co/JHNfJ9tWOO #TRICON
6:13pm March 9th 2016 via Hootsuite
Daber: Rearrangement: DNA doesn't form chimeric protein; intact protein near an active promoter. Vs. Fusions: new chimeric protein #TRICON
6:11pm March 9th 2016 via Hootsuite
Daber: Quant: could replace IHC; Qual: fusion transcripts, EGFR VIII (exons 2-7 deleted), MET exon 14 skipping #TRICON
6:10pm March 9th 2016 via Hootsuite
Daber: Applications: discovery of novel fusion partners; deep population sequencinig (NTRK fusions across tumor types) #TRICON
6:08pm March 9th 2016 via Hootsuite
