Donald Trump's US election win stuns scientists : Nature News https://t.co/vDvRslRXwn
6:00pm November 9th 2016 via Hootsuite
Q: Hard to call areas for the background model? Ionescu-Zanetti: Restricted search spaces can be little or large #AMP2016
3:55pm November 9th 2016 via Hootsuite
Ionescu-Zanetti: Table points out FP/10kb not reported via Guardant, HloPlex. FMI has 0.02 reported, but AF 2-5%. Spotlight 59 0.2% #AMP2016
3:51pm November 9th 2016 via Hootsuite
Ionescu-Zanetti: A statistical test on a per-base basis. Has detailed comparison chart against FMI, Guardant, QIAseq, HaloPlex #AMP2016
3:50pm November 9th 2016 via Hootsuite
Ionescu-Zanetti: Technical replicates - 'jackpot errors' easily found, eliminated. Stochastic ones use a background model #AMP2016
3:45pm November 9th 2016 via Hootsuite
Ionescu-Zanetti: Of a given # of calls, some true pos, other recurring artifacts, others stochastic. Uses replicate samples #AMP2016
3:44pm November 9th 2016 via Hootsuite
Ionescu-Zanetti: Describes method called ERASE-seq. Recent press release: https://t.co/cZA71qSrsx Early-amplification errors #AMP2016
3:42pm November 9th 2016 via Hootsuite
Ionescu-Zanetti: cfDNA in Streck BCT, n=8; use QIAamp circ NA kit; 10ng to Accel-Amplicon #AMP2016
3:38pm November 9th 2016 via Hootsuite
Ionescu-Zanetti: UMI reduce errors, but high depth required; sensitive to 1%, but spec goes down if sens at 0.1-0.5% #AMP2016
3:37pm November 9th 2016 via Hootsuite
Ionescu-Zanetti: Below 1%, both cfDNA and CTCs are affected by total number of copies. Molecular barcodes - they don't #AMP2016
3:36pm November 9th 2016 via Hootsuite
Cristian Ionescu-Zanetti (CTO Fluxion) Ultra-sensitive sequencing for liquid biopsy samples #AMP2016
3:35pm November 9th 2016 via Hootsuite
Q: Large number w/muts? Martignetti: Have looked again, but no evidence. Of 100, expect 5-10. Of the 90, one has returned to-date #AMP2016
3:31pm November 9th 2016 via Hootsuite
Martignetti: Validated with Sanger (AF >10%) or ddPCR (AF <10%). Shows photo of a slide bank from 1 pt: 1mm marked by hand #AMP2016
3:28pm November 9th 2016 via Hootsuite
Martignetti: Other 100 pts - polyps, normal, fibroids. 51 women had driver muts in lavage 'at rather high freq'. 1%-30%; ave 3% #AMP2016
3:25pm November 9th 2016 via Hootsuite
Martignetti: Needed both cell pellet and cfDNA as not consistent. Tumor seq showed same muts 100% #AMP2016
3:23pm November 9th 2016 via Hootsuite
Martignetti: Of 107, 7 had endometrial ca. Stains, 1mm-2mm lesions, each had detected mutations in lavage. PTEN, ARID1A, PIK3CA #AMP2016
Martignetti: Prospective study: 40-59yo, shows table '15 ref https://t.co/FEggROrtox #AMP2016
3:21pm November 9th 2016 via Hootsuite
Martignetti: Using a panel - 5 genes, can they use a MDx on lavage fluid? 107pts for Hysteroscopy, split w/lavage vs pathology #AMP2016
3:19pm November 9th 2016 via Hootsuite
Martignetti: Outliers: Type I w/recurrence. #TCGA discovered 4 subtypes. '13 ref https://t.co/O4041mlkXs #AMP2016
3:18pm November 9th 2016 via Hootsuite
Martignetti: A subjective test with natural history. Type I/II: will determine surgical plan, follow-through. #AMP2016
3:16pm November 9th 2016 via Hootsuite
Martignetti: No useful Dx; use OR-based hyteroscopy. 15% cases inadequate tissue, subjectivity. Cell-type and grade. #AMP2016
3:15pm November 9th 2016 via Hootsuite
Martignetti: Developed ctDNA for surveillance, now want a screening tool. Symptoms: non-menstrual bleeding; dysuria, pain #AMP2016
3:14pm November 9th 2016 via Hootsuite
Martignetti: Endometrial ca most common gynecological malignancy; >55K new cases, >10K deaths. Will be #3 cancer site in females #AMP2
3:13pm November 9th 2016 via Hootsuite
John Martignetti (Icahn SOM at Mt Sinai NY) Mutatio profiling of uterine lavage fluid - early-stage endometrial cancers #AMP2016
3:12pm November 9th 2016 via Hootsuite
Harkins: Able to do overlapping amplicons in the same tube; thus whole TP53 gene coverage. #AMP2016
3:11pm November 9th 2016 via Hootsuite
Harkins: cfDNA (PCR-free) high-AT is lost. Shows some Accel-amplicon uniformity data across sample types. Has BRCA1/2 now #AMP2016
Harkins: Shows data w/PCR-free 2S lib prep. ccfDNA 'is not whole-genome'. Shows Picard plots based on GC% at 100ng, 10ng, cfDNA #AMP2016
3:09pm November 9th 2016 via Hootsuite
Harkins: Shows data from chemagic 360 (PerkinElmer). #AMP2016
3:08pm November 9th 2016 via Hootsuite
Harkins: HMW contam. means higher 247/115 ratio (since HMW >0.9). Recommend Streck BCT, isolation within 72h #AMP2016
3:07pm November 9th 2016 via Hootsuite
Harkins: cfDNA normal = 0.38 ratio; ctDNA = 0.22. FFPE at different fixation times - 6h =0.5, 24h = 0.4 #AMP2016
3:06pm November 9th 2016 via Hootsuite
Harkins:. Lower input and more damage go hand-in-hand. For ccfDNA - quality assessed via Alu247/Alu114 to calc 'integrity score' #AMP2016
3:05pm November 9th 2016 via Hootsuite
Tim Harkins (Swift Biosciences) Reviews their portfolio - 2S (ds apps); accel-amplicon (PCR from 10ng), Methyl-Seq, and 1S (ssDNA) #AMP2016
3:04pm November 9th 2016 via Hootsuite
Krishnan: 5ng total RNA, even 1ng possible with Ultra II; in early-access. #AMP2016
2:29pm November 9th 2016 via Hootsuite
Krishnan: Use RNaseH method to degrade rRNA (hybrid duplexed degraded). 2h workflow, 10' hands-on <1% left #AMP2016
2:22pm November 9th 2016 via Hootsuite
Keerthana Krishnan (Dev Scientist NEB) Will be talking about RNA library challenges. rRNA contam, low input, sub-optimal quality #AMP2016
2:19pm November 9th 2016 via Hootsuite
Q: UMI and software? Barry: Have a version of Picard to mark duplicates #AMP2016
2:18pm November 9th 2016 via Hootsuite
Q: Tried lower input? Barry: 100ng input rec'd. Have good understanding of performance. Can go down to 10ng, but for cf 100ng #AMP2016
2:17pm November 9th 2016 via Hootsuite
Barry: Since pre-synthesized, can relatively quickly put it together. Currently in early-access. #AMP2016
2:15pm November 9th 2016 via Hootsuite
Barry: Showed 2% to 100% MAF linearity; shows sensitivity, spec and uniformity compared to others. Have baits for 450 genes, 1MB #AMP2016
2:14pm November 9th 2016 via Hootsuite
Barry: Bases <33x of the mean is 1.5%; side-by-side competition is much higher (measuring specificity of capture) #AMP2016
2:13pm November 9th 2016 via Hootsuite
Barry: Target size optimized for 2x75 PE (ILMN). Shows % mappability across fresh-frozen, cell-free and FFPE samples #AMP2016
2:11pm November 9th 2016 via Hootsuite
Barry: Shows tiling across KIT; nice coverage plot. Able to scale from 17kb to 360kb targets w/o alignment and % on-target loss #AMP2016
2:10pm November 9th 2016 via Hootsuite
Barry: Same-day sample-to-seq; no lo-vol transfers; no heated buffers. UMI is 12bp. #AMP2016
2:08pm November 9th 2016 via Hootsuite
Barry: They do hyb upfront; then convert to library. Hyb targets both strands at 3' end; after strepavidin pull-down bait stays on #AMP2016
2:07pm November 9th 2016 via Hootsuite
Barry: Needs: low input (<100ng); FFPE / blood / cfDNA); UMI's, sens. spec. uniformity; flexibility; automatable #AMP2016
2:05pm November 9th 2016 via Hootsuite
Andrew Barry (New England Biolabs, product manager) A unique method for target enrichment: NEBNext Direct #AMP2016
2:04pm November 9th 2016 via Hootsuite
SeraCare Life Sciences collaborates with ArcherDX on RNA Fusion Reference Standards PR Newswire https://t.co/77IfI656Oa #AMP2016
1:34pm November 9th 2016 via Hootsuite
Q: Amount of blood? (Personal GenomeDx): 2 or 3 10mL tubes. #AMP2016
1:33pm November 9th 2016 via Hootsuite
(Personal GenomeDx): EDTA can be kept at 4C; equivalent results to Streck BCT at RT. Hospitals may not be able to fractionate #AMP2016
1:32pm November 9th 2016 via Hootsuite
(Personal GenomeDx): Showed effect of storage - EDTA increases at RT, but 4C is fine over 2d. But EDTA 4C shows dec in MAF #AMP2016
1:26pm November 9th 2016 via Hootsuite
