Archive of Saved Favorite Tweets from 9 Nov. '15 to 6 Aug. '16 [twitter0mg]

  1. Is the 2016 Presidential Election about and driven by: partisanship, fear, trust, or anger?
  2. @markgerstein @carlzimmer @statnews closed humans isn't a club, any more than 'didn't have-pizza' is a club.
  3. Thanks @aayushraman It was a team effort by @markgerstein lab and @carlzimmer was very kind to give us this opportunity.
  4. @matthewherper @carlzimmer @statnews Most interesting tidbit was that Gerstein said he would not have had courage to do this himself!
  5. About the series: Watching my genome flow into Gerstein's computer made me a little giddy. ... Scientists hav...  http://bit.ly/29BDpHg 
  6. Congratulations Juno! You made it in a win for computation. Here's some Jupiter info from Nautilus to celebrate  http://m.nautil.us/issue/22/slow/jupiter-is-a-garden-of-storms 
  7. Gem from @yaringal's poster "Uncertainty is everywhere. Train a model to recognise dog breeds; someone will try to classify a cat" #icml2016
  8. Awesome analysis of LP relaxation tightness in structured prediction by Ofer Meshi #icml2016 https://t.co/0M589YVHvE
    Awesome analysis of LP relaxation tightness in structured prediction by Ofer Meshi #icml2016 pic.twitter.com/0M589YVHvE
  9. Classifying videos with rank-pooling to capture dynamics in sequences : could we use it for music audio ?  http://jmlr.org/proceedings/papers/v48/fernando16.pdf  #icml2016
  10. Very cool #icml2016 workshop! Picking the "right" tasks/data is as important choosing the "right" models, I think.  https://twitter.com/nycmedialab/status/744905723872673793 
  11. Best paper award at #icml2016 : dueling network architectures for deep reinforcement learning by @DeepMindAI  https://goo.gl/UFX3VF 
  12. My slides #icml2016: online learning + random subsampling for fast matrix factorization. Thanks to the audience !  http://amensch.fr/docs/presentations/icml_presentation.pdf 
  13. Almost 90% increase in #icml2016 attendance. Being a first timer I am one of the contributors to that number 😉 https://t.co/6XCMgm5GGo
    Almost 90% increase in #icml2016 attendance. Being a first timer I am one of the contributors to that number 😉 pic.twitter.com/6XCMgm5GGo
  14. To identify rare variants, the challenge is numbers - need to analyze millions of people. @markgerstein #GBMseminar
  15. “One of the key ways we grapple w/huge amount of data is prioritization. One way to do this is to focus on the rare variants.” @markgerstein
  16. PersonalGenomics:Managing Exponential DataScaling through Prioritizng High-Impact Variants @markgerstein #GBMSeminar https://t.co/51niMcmw2s
    PersonalGenomics:Managing Exponential DataScaling through Prioritizng High-Impact Variants @markgerstein #GBMSeminar pic.twitter.com/51niMcmw2s
  17. “The big thing happening today is the big amount of private data being collected –>new challenge in terms of access & privacy" @markgerstein
  18. Prepping for highly anticipated talk from @markgerstein for GBM series @IcahnInstitute @IcahnMountSinai Davis in Hess @ 1pm
  19. HA Orr in the NY Review of Books, on DNA, theory, and experiment in biology. I wish I had written this.  http://www.nybooks.com/articles/2016/06/09/dna-power-beautiful-experiment/  via @nybooks
  20. Presenting Projects from UCB and Big Data in Biomedicine @GTCbio
  21. Great to hear Jeff Schloss/NIH reflect on the past 20 years of remarkable advances in DNA Sequencing Technology - truly amazing @GTCbio
  22. Second best. Best was President #Obama telling me my family is gorgeous & sending over #M&Ms for the little ones. https://t.co/A5xFYj3pds
    Second best. Best was President #Obama telling me my family is gorgeous & sending over #M&Ms for the little ones. pic.twitter.com/A5xFYj3pds
  23. Only the Personal Genome Project attack in Table attempted re-identification for everyone who had data in question.  https://twitter.com/markgerstein/status/730215312801583105 
  24. But the methods issues in their purported "proof of concept" attack shed little light on true Re-ID feasibility.  https://twitter.com/markgerstein/status/728406501191254016 
  25. Congratulations to my colleague and next-door lab neighbor Fred Sigworth for his election to the National Academy of Science! 🎉🎉🎉🎉🎉🎉🎉
  26. @AMathelier @markgerstein 0.3 is arbitrary-but chosen to balance between chipseq dataset removal and not overcalling asb snvs too much
  27. @AMathelier @markgerstein ASE! Sry misread! Not here. Explored some in previous papers eg Alleleseq. Agree-will be nice to see coordination.
  28. @cjieming @markgerstein The section doesn't look at ASE. Would be good to know if strong binding disruption links to strong ASE or not.
  29. @AMathelier @markgerstein thx for the interest! Yes we did - it's in paper, section "AS variants in TF-binding motifs affect TF occupancy"!
  30. @markgerstein Did you look at correlation between motif disruption strength and ASE strength? 1/2
  31. @markgerstein Really nice work! Using betabinomial is very accurate. But why 0.3 as thr for chipseq? Other tested? 1/2
  32. @MarkGerstein These might be some of the hacks you were thinking of? None against data that was #HIPAA de-identified https://t.co/Cx2ykhZNtJ
    @markgerstein These might be some of the hacks you were thinking of? None against data that was #HIPAA de-identified pic.twitter.com/Cx2ykhZNtJ
  33. De-identification can always produce data w/ low Re-ID risks. <=But may come at expense of reducing dimensionality.  https://twitter.com/markgerstein/status/726431688730009601 
  34. Ready to discuss. Can you list the hack examples? Expect you'll find no statistical disclosure control was used.  https://twitter.com/markgerstein/status/726431688730009601 
  35. Allele-specific transcription factor binding in liver and cervix cells unveils many likely drivers of GWAS signals.  http://dlvr.it/LBYzj0 
  36. @reactome love the expression atlas integration, but fyi you're sporting a couple left-handed helices #biopetpeeves https://t.co/QpBkbufyTO
    @reactome love the expression atlas integration, but fyi you're sporting a couple left-handed helices #biopetpeeves pic.twitter.com/QpBkbufyTO
  37. @markgerstein somehow I imagine LMMs or GAMs might take a bit too long ;-). Nice catch, thanks for pointing it out!
  38. A uniform survey of allele-specific binding and expression over 1000-Genomes-Project individuals: Nature Commu...  http://bit.ly/1VmgIZ1 
  39. A uniform survey of allele-specific binding and expression over 1000-Genomes-Project individuals.: A uniform s...  http://1.usa.gov/1re8kOO 
  40. A uniform survey of allele-specific binding and expression over 1000-Genomes-Project individuals.  http://dlvr.it/L5WQLK 
  41. A uniform survey of allele-specific binding and expression over 1000-Genomes-Project individuals.  http://dlvr.it/L5WKRS 
  42. Researchers annotate elements w/ different activity in mother's & father's personal genomes  http://bit.ly/1Nx53P1  @markgerstein lab
  43. Research note: Don’t forget parents when celebrating a personal genome  http://ift.tt/1WASgCw 
  44. @markgerstein A brill read! I read this paper last week and really enjoyed it! Great that it separates interior/exterior critical residues.
  45. A Visionary Project Aims for Alpha Centauri, a Star 4.37 Light-Years Away  http://nyti.ms/1qmayLd 
  46. when #privacy & public health collide. Dr. Laurie Glimcher on patient consent regulation & cancer research:  http://on.wsj.com/1W7pLfj  via @WSJ
  47. Eli Lilly's Good Cholesterol Drug Went Bad. Here's What That Means For Pharma  https://lnkd.in/bcF99Tu 
  48. Can anyone tell me what this image is? April fools? DNA acting like RNA? Great sharing story @pollyp1 pic.twitter.com/UZRBAKn4j2
  49. It’s April 1st and time to hide from the internet.
  50. @Piwdb @markgerstein always good to read Mark's paper, the general tendency being to shoot data w/o thinking of what to do with it..
  51. @markgerstein @ENCODE_NIH Nice figure. Who are the poor souls on the circumference of the wheel?
  52. @markgerstein @ENCODE_NIH Awesome! Congratulations Prof. Your paper can easily baseline into #QIS Dynamic interpretations are hard, hence...
  53. The US as it might appear with all the proposed states. #dataviz

@USouthFlorida would be in trouble w/ this divide! https://t.co/k0u3swdtmm
    The US as it might appear with all the proposed states. #dataviz @USouthFlorida would be in trouble w/ this divide! pic.twitter.com/k0u3swdtmm
  54. @markgerstein interesting that you mention - I just installed @Lightroom on my phone - cannot convince myself it's better than google :-(
  55. Large-scale atlas of microarray data reveals the distinct expression landscape of different tissues in Arabidopsis. - PubMed - NCBI …
  56. The real cost of sequencing: scaling computation to keep pace with data generation | Genome Biology | Full Text -  http://go.shr.lc/1RFkYSz 
  57. Large-scale atlas of microarray data reveals the distinct expression landscape of different t...  http://www.ncbi.nlm.nih.gov/pubmed/27015116 
  58. Large-scale atlas of microarray data reveals the distinct expression landscape of different tissues in Arabido...  http://1.usa.gov/1pOmRQz 
  59. Large-scale atlas of microarray data reveals the distinct expression landscape of different tissues in Arabidopsis.  http://dlvr.it/KtmRTQ 
  60. Fig 4. にUS51大学のゲノム配列解析に絡んだfaculty position数が出てますね。日本にも51大学くらいありますが、1/5以下じゃないだろうか(数えてません)。 http://genomebiology.biomedcentral.com/articles/10.1186/s13059-016-0917-0 
  61. 計算機の速度と配列解析の歴史。//The real cost of sequencing: scaling computation to keep pace with data generation  http://genomebiology.biomedcentral.com/articles/10.1186/s13059-016-0917-0 
  62. The real cost of sequencing: scaling computation to keep pace with data generation  http://buff.ly/1XQJ9fa 
  63. @nils_gehlenborg @minouye271 @markgerstein @CellSystemsCP My cat’s hairballs look nothing like any large network. Much more mucus and such.
  64. Sorry @minouye271 @markgerstein @CellSystemsCP, but networks =/= hairballs. Poor network visualizations => hairballs.
  65. @markgerstein Yeah. Thanks God there is no Nobel for them (sorry Werner :-) How many of your former people are now in that category?
  66. The real cost of sequencing: scaling computation to keep pace with data generation - Genome Biology  http://nzzl.us/RR91dAV 
  67. The real cost of sequencing: scaling computation 2 keep pace w/ data generation  http://ht.ly/ZTB24  via @GenomeBiology #bioinformatics
  68. The real cost of sequencing: scaling computation to keep pace with data generation  http://ow.ly/ZTuYi 
  69. Interesting article on bioinformatics' data generation and scaling computation to keep pace  http://goo.gl/9CQZGQ  #genomebiology
  70. The real cost of sequencing: scaling computation to keep pace with data generation...  http://fb.me/4Kf9cq1Cr 
  71. Drugmakers tackle hearing loss https://t.co/Zj6swGONiv Plus: cool interactive on how hearing works @lisamjarvis https://t.co/1uqn0P5SPK
    Drugmakers tackle hearing loss  http://ow.ly/ZHJTP  Plus: cool interactive on how hearing works @lisamjarvis pic.twitter.com/1uqn0P5SPK
  72. @markgerstein Yes, indeed. I could read it. Nobel shouldn't be tampered with, in my op. 1M is still a lot of money for us :-)
  73. Everyone thinks Lee Sedol won but did you look at the black pieces pattern??? #AlphaGo #AI #Skynet https://t.co/XXzjN5oSFf
    Everyone thinks Lee Sedol won but did you look at the black pieces pattern??? #AlphaGo #AI #Skynet pic.twitter.com/XXzjN5oSFf
  74. Interesting eg: Distorting real world Geo for network node embedding/connecting arcs can help us better understand  https://twitter.com/edwardtufte/status/708660827788853249 
  75. Going beyond geek chic — CeBIT: Kaleidoscopic mirrors reflect slogans and fair visitors at the stand of German...  http://bit.ly/1TyB5Bh 
  76. Going beyond geek chic — CeBIT: ... last season's shoes to fashion week, failing to keep up with the latest de...  http://bit.ly/1TyuuXA 
  77. Going beyond geek chic — CeBIT - San Francisco Chronicle: Going beyond geek chic — CeBITSan Francisco Chronicl...  http://bit.ly/1UjAhyY 
  78. FRESH in Pubmed: My "QTL Identify Functional Noncoding Variation in Cancer." @PLOSGenetics  http://goo.gl/GcERY3  Incl.Funseq @markgerstein
  79. Check out our science section and learn about VarSim, A high fidelity simulation and validation framework for...  http://fb.me/2A2N6u3Qr 
  80. Learn about BreakSeq2: a fast and accurate junction-mapping approach for detecting structural variants  http://bit.ly/1WSw2tc 
  81. Attention Twitter: @richard_durbin has tweeted for the first time today. Welcome Richard!
  82. Fascinated to learn Jennifer Doudna, of CRISPR fame, and Erin O’Shea, named 1st woman prez of HHMI today, both married their grad students
  83. AOP: What’s the risk of being identified by independent data on phenotype & genotype? ($)  http://bit.ly/1PMs6Fl 
  84. As genomic information grows, so do privacy risks — as shown by work in @markgerstein lab  http://bit.ly/1nIorm0 
  85. The DNA in the poster is spiraling in the wrong direction. Ignorance of molecular biology is poor science.  https://twitter.com/NatureNews/status/693520431417008129 
  86. Epic fine-mapping work on a schizophrenia GWAS hit identifies expression of C4A gene as a likely causal mechanism  http://www.nature.com/nature/journal/vaop/ncurrent/full/nature16549.html 
  87. Got the 1st copy of "Genomics & Personalized Medicine" off the press, signed by Mike Snyder himself! #PMWC16  https://www.amazon.com/dp/0190234768/ref=cm_sw_r_cp_awd_ai4PwbG5CKADJ 
  88. Why students develop an aversion to mathematics — and how teachers can help change their minds  http://hvrd.me/XvMsN 
  89. Mike Snyder now wearing 3 smart watches and a continuous glucose monitor! Probably needs more skin real estate for wearables! #PMWC16
  90. @BioMickWatson @iddux Awesome. Checked "Falling down stairs" (how my parents separately got head injuries) - correlates with cost of bananas
  91. Welcome to "Data Science, Learning, and Applications to Biomedical & Health Sciences" w/ @Rutgers_Newark & @DSI_Columbia. Follow #BDHealth
  92. In Sunday's @NYTmag, how a lawyer discovered that a toxin tested in secret by DuPont has ended up in our blood:  http://goo.gl/ZlOLc3 
  93. I'm going to choose to think about this finding as one way to increase the diversity of my gut microbiome. : )  https://twitter.com/markgerstein/status/679005719711129600 
  94. Thanks to @vsbuffalo for his help setting up a html version of my popgen notes  http://cooplab.github.io/popgen-notes/  making use of gh-pages+pandoc