Lincoln #AGBT19 Table 2: whole list of recommendations. https://t.co/mjrmvr0QzL
10:23am March 2nd 2019 via Twitter Web Client
Lincoln #AGBT19 Lays out how they tried to reduce overfitting.
10:21am March 2nd 2019 via Twitter Web Client
Lincoln #AGBT19 Imbalanced dataset: TPs vastly outnumber FPs. Many low-qual calls are TPs. The cost of a missed TP is great.
10:18am March 2nd 2019 via Twitter Web Client
Lincoln #AGBT19 Qual metrics, allele balance, read depth, content, etc. Many FP's have 'good' allele balance (!) Ma… https://t.co/P8abDNM8gW
Lincoln #AGBT19 Clin confirmation data from 80K pts, all five NIST samples, on- and off-target var calls, two labs.… https://t.co/0Dx9fKwWDy
10:17am March 2nd 2019 via Twitter Web Client
Lincoln #AGBT19 Data sets ize and diversity matters. Mu et al - >100x and AF >40%. The new study: 29 FP indels and 7 FP SNVs !
10:16am March 2nd 2019 via Twitter Web Client
Lincoln #AGBT19 How to rigorously est upper criteria? Many papers '15 ref https://t.co/L8JIIogluR
10:14am March 2nd 2019 via Twitter Web Client
Lincoln #AGBT19 Low thresholds for sensitivity, and a high threshold for high specficity. Collab with Matt Levo Har… https://t.co/86zibx3VBy
10:13am March 2nd 2019 via Twitter Web Client
Lincoln #AGBT19 Six different platforms! PacBio, short amplicon, LR-PCR, Sanger. Variants called from NGS - the one… https://t.co/ZNAlDMjtRZ
10:12am March 2nd 2019 via Twitter Web Client
Lincoln #AGBT19 Optimize for sens, enforce spec by an orthogonal method. and then clinical reporting, pt mgmt. Addi… https://t.co/1qw9aid9Qx
10:11am March 2nd 2019 via Twitter Web Client
Lincoln #AGBT19 Challenges - large % (9-19%) are technically challenging for NGS (large indels, single-exon CNVs, S… https://t.co/TLQUbJgssT
10:10am March 2nd 2019 via Twitter Web Client
Lincoln #AGBT19 Just published in J Mol Diag https://t.co/O4U6hsi4OX
10:09am March 2nd 2019 via Twitter Web Client
Stephen Lincoln (Invitae) #AGBT19 A rigorous study of the need for Sanger confirmation in clinical genetic testing
Q: Using blood - how often will it work or not? Montgomery #AGBT19 A bonus slide: about 66%-70% all known disease g… https://t.co/m8jWoOltdH
10:08am March 2nd 2019 via Twitter Web Client
Montgomery #AGBT19 'The next generation of genome interpretation tools will integrate personal functional genomics.'
10:05am March 2nd 2019 via Twitter Web Client
Montgomery #AGBT19 RNA-seq on diverse cases yields a 7.5% diagnostic rate. Add'l 16.7% with a 'strong candidate'. G… https://t.co/ctl4vEEckh
10:04am March 2nd 2019 via Twitter Web Client
Montgomery #AGBT19 Preprint https://t.co/P6Ygdvfixz for RNA-seq in 3 rare disease cohorts. RNAseq from 143 whole bl… https://t.co/FWFMHKrt1d
9:57am March 2nd 2019 via Twitter Web Client
Montgomery #AGBT19 RNA-seq helps '17 ref https://t.co/5uW5WhFO7n Can help ID rare disease var's and genes.
9:56am March 2nd 2019 via Twitter Web Client
Montgomery #AGBT19 Est of diagnosis of WES: 30%-50% success rate. Reviewed '19 https://t.co/4cCzG3Nb3N
9:55am March 2nd 2019 via Twitter Web Client
Montgomery #AGBT19 A large challenge - what do these singleton mutations do? Association analyses fail when var's are rare.
9:37am March 2nd 2019 via Twitter Web Client
Montgomery #AGBT19 WGS reveals rare human variation. From gnomAD - 15708 geomes, 120M singleton variants (!)
Stephen Montgomery (Stanford Univ) #AGBT19 Functional genomics for rare variants and rare diseases
9:36am March 2nd 2019 via Twitter Web Client
Pollard #AGBT19 ShapeMF Gibbs sampling algorithm. 19 ref Cell Systems https://t.co/MUbeEkgXn5
9:30am March 2nd 2019 via Twitter Web Client
Pollard #AGBT19 Shape motifs extend sequence motifs: shape and sequence sites can overlap, extend, have conserved spacing.
9:28am March 2nd 2019 via Twitter Web Client
Pollard #AGBT19 Shows data from 6 ChIP and SELEX motifs - DNA binding proteins recognize motifs.
9:26am March 2nd 2019 via Twitter Web Client
Pollard #AGBT19 Shows interesting plot of 'roll' function. Applied to in-vivo (ENCODE ChIP) and in vitro.
9:24am March 2nd 2019 via Twitter Web Client
Pollard #AGBT19 Idea: translate human genome from nt's to a quantitative description of structure at each time. Hel… https://t.co/hntLanLMVa
9:22am March 2nd 2019 via Twitter Web Client
Pollard #AGBT19 DNA binding proteins have similar seq motifs yet bind different sites. And DNA structure of these s… https://t.co/j42mfryutW
9:19am March 2nd 2019 via Twitter Web Client
Pollard #AGBT19 DNA binding proteins: across 110 TF's in ENCODE: 30% of the top peaks 'lack a sequence motif'. Comm… https://t.co/Q4sz22uPvn
9:18am March 2nd 2019 via Twitter Web Client
Pollard #AGBT19 Looking at LD and chromatin maps: no correlation, b/c LD is relatively short whereas chromatin inte… https://t.co/TXlyOJWFc6
9:14am March 2nd 2019 via Twitter Web Client
Pollard #AGBT19 Concl: non-coding mutation scoring tools should be chromatin structure-aware.
9:11am March 2nd 2019 via Twitter Web Client
Pollard #AGBT19 See that selection correlates with function. In autism and cancer, not observed - enhancer hijackin… https://t.co/nkaEP5Bljf
9:10am March 2nd 2019 via Twitter Web Client
Pollard #AGBT19 Looking at deletions - large effect size, boundaries are clear. They've detected (using Hi-C data,… https://t.co/9SpeZjFdY6
9:07am March 2nd 2019 via Twitter Web Client
Pollard #AGBT19 Chromatin capture, TADs, boundary elements (BEs); how often are mutations affecting structure removed from the genome?
9:06am March 2nd 2019 via Twitter Web Client
Pollard #AGBT19 Thinking of DNA as a 3D structure; chromatin structure meets population genetics. 3 stories - chromatin boundaries.
9:04am March 2nd 2019 via Twitter Web Client
Pollard #AGBT19 If a particular structure is functional, mutations that change it would be deleterious. 'A broader… https://t.co/nv29gVsU00
Katherine Pollard (Gladstone Institute UCSF) #AGBT19 A population genetic view of human chromatin organization
9:03am March 2nd 2019 via Twitter Web Client
New post: The age of mRNA isoform discovery at #AGBT19 https://t.co/kcprNjb0zX
7:23am March 2nd 2019 via WP to TWTR on Yuzuki.org
@DrEva24374996 Challenge accepted! Will be sure to swing by today at #AGBT19.
6:03am March 2nd 2019 via Hootsuite Inc. in reply to DrEva24374996
Q: What is 'novel' - new transcribed regions? Underwood #AGBT19 Low expressed, usually degraded. 3' UTR may have re… https://t.co/yN2Vly5Uej
9:34pm March 1st 2019 via Twitter Web Client
Underwood #AGBT19 Conclude: DropSeq can produce full-length cDNA with single cell barcoding info. New biology awaits!
9:31pm March 1st 2019 via Twitter Web Client
Underwood #AGBT19 Post-transcriptional gene regulation display: an intron in a 3' UTR of EIF4h, triggering non-sens… https://t.co/gmMhEgM326
9:27pm March 1st 2019 via Twitter Web Client
Underwood #AGBT19 Splice junctions - 87% in Gencode v29; 13% were novel. Looking at the 13%: 3.8% human dev cortex… https://t.co/Be1Qnsq8B5
9:25pm March 1st 2019 via Twitter Web Client
Underwood #AGBT19 Sequel II (8M) datasets - reviews how the data was cleaned up, genes detected per cell 729/cell.
9:22pm March 1st 2019 via Twitter Web Client
Underwood #AGBT19 Initial data look good with the cell mixes. Now with cell organoids: '19 Cell https://t.co/brk6XR9hCd
9:19pm March 1st 2019 via Twitter Web Client
Underwood #AGBT19 Standard experiment - mixing cell lines 3T3 (mouse) and 293 (human). Full-length cDNA PCR, then P… https://t.co/6Yd4LvpFJa
9:16pm March 1st 2019 via Twitter Web Client
Underwood #AGBT19 DropSeq for full-length cDNA using oligos on solid beads with adjacent template switching. DropSeq to feed into PacBio.
9:15pm March 1st 2019 via Twitter Web Client
Underwood #AGBT19 Turning up the information content; from gene expressoin to isoform expression. 3' tagging (10X,… https://t.co/A8dt7abAv7
9:14pm March 1st 2019 via Twitter Web Client
Underwood #AGBT19 Working with Euan Eichler's group, also working with Dolomite Bio's group (blog post… https://t.co/q5xjGA2ZFR
9:13pm March 1st 2019 via Twitter Web Client
Jason Underwood (PacBio) #AGBT19 “Single cell isoform sequencing (scIso-Seq) identifies novel full-length mRNAs and… https://t.co/aSelquX5cb
9:12pm March 1st 2019 via Twitter Web Client
