#ampeurope2018 Dalton: Pt enrichment for clinical trials - recent publ https://t.co/bDtaYNlOFJ
10:18am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Dalton: For a given individual going to a physician: included in a cohort that looks just like them.
10:14am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Dalton: Data-sharing is collaboration: but an agreement that that is the goal. Then out to the point of care.
10:13am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Dalton: If you are sharing pt data, they need to know. Vision of breaking down silos; standard proto… https://t.co/HNKCeIkivs
10:12am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Dalton: Uses Senteion, and DNANnexus. Data warehousing by cloudicity, and API by click.
10:11am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Dalton: Started at HudsonAlpha, recently added TGEN for genetic sequencing, same SOPs, platforms, data handling
10:10am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Dalton: Cancer registry, with classic demographics. Common data elements agreed-upon. Now almost 50K new pts/y
10:09am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Dalton: Now WES at 300x; RNASeq 100M; Germline at 100x. Data comes back to the pt, and a federated data sharing
10:07am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Dalton: Want to anticipate need for biomarker-driven clinical trials 'before they progress to create options'
10:06am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Dalton: 16K legacy pts w/ gene exp profiles; have started w/pts with greatest need.
10:05am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Dalton: A system of continuous learning and ability to anticipate need. In 12y: have legacy data w/microarray
10:03am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Dalton: Power is cohort surveillance; virtual 'in silico communities'. Longitudinal clinical data
10:02am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Dalton: TCC protocol is same handling of tissue, clinical and molecular data. M2GEN facilitates the… https://t.co/fTn3ZdpYOz
10:01am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Dalton: Focus on data utilization. Design approach to jobs to address. Many stakeholders.
#ampeurope2018 Dalton: Now 211K pts consented to-date. Lessons: need for scale, vol, velocity, variety, and veracity. Anticipate needs.
10:00am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Dalton: Allow access to entire medical record following them for life, and permission to re-contact.
9:59am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Dalton: Dev. markers to predict need in order to prevent; pts consent to donate tissue for storage, future analysis...
9:58am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Dalton: M2GEN founder. ORIEN now has 17 members. "Total Cancer Care". ID's needs of pt and families, evidence-based approach
9:57am May 2nd 2018 via Twitter Web Client
#ampeurope2018 William Dalton (Moffitt Cancer Center Tampa FL USA) Oncology Research Information Exchange Network (ORIEN)
9:56am May 2nd 2018 via Twitter Web Client
Q: Clin decision-making support in EHR? #ampeurope2018 Hoffman: Blockchain not mature / ready. Storing genomic data… https://t.co/4EQUdNDnAf
9:54am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: CYP450 variation, now crowdsourced CYP2D6, 2D9, and other var's entered into https://t.co/qclgTn45rz
9:51am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: To contribute as subject matter experts (SMEs), boundary checking, thoughtful defaults
9:50am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: What to do: implementation science. Data science teams 'have to be interdisciplinary'. SEM and Technical expert.
9:49am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: Few labs discuss release of 2nd-ary findings to family.
#ampeurope2018 Hoffman: 18 academic consnet forms, and worked through consent. https://t.co/0e8KsKkVVL Complexity was 10.8 and rec'm was 8
9:48am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: Maintaining curation harder and harder. Fixed codes w/sequencing, cytogenetics: something unanticipated.
9:47am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: Clinical Bioinformatics Ontology: '05 https://t.co/HDE0VAGa2m As the field has become more complex...
9:46am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: Shows cahrt of HbA1c codes for different sickle-cell conditions in ICD-9 and ICD-10.
9:45am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: LOINC, SNOMED, ICD-10 are terminologies. Other option is to formalize a syntax. More often in MDx, say HGV
9:44am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: interoperability: NA, Sod, Sodium supposed to mean the same thing. E. coli O157:11 variant with a like concept
#ampeurope2018 Hoffman: Truman: 3x the rate of A1c test on SS-affected, room to improve by making a change in their system 'popup alert'
9:40am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: Ethnic demo: conclude that Truman Medical (their site) admits less frequent 'biracial' intake code
9:39am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: 863 facilities, 734M medication orders, 489M diagnoses. ID'd 41K SS-affected (40% of the US-affected)
9:38am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: They have a CDC grant to use large data warehouses to improve quality. 63M pts, 4.3B lab results.
9:37am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: They have aggregate EHR data, De-ID, Map, normalize into 'Health Facts' from Cerner.
9:35am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: Est 100K individuals in the US affected. Physicians may forget how HbA1c test works: may no… https://t.co/8w1r9brBFE
#ampeurope2018 Hoffman: Big Data with common genetic disease: sickle cell, HbA normal, HbSS 1/13 Af Americans carriers
9:34am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: Most data must be manipulated before feeding algorithm; decisions need to be rigorously documented.
9:33am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: We imagin a well-oiled machine, but it's the wizard of Oz behind the curtain (GIFs shown!)
#ampeurope2018 Hoffman: Big data my have unk origin, out of context, inappropriate level of rigor, and hype.
9:32am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: Espt with experimental methology nuances. Experts can't necessarily ID the correct result.
9:31am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: "Big data analysts don't always understand little data effects". They don't have domain kno… https://t.co/Qd0NxuTlSc
#ampeurope2018 Hoffman: Pediatric genomics at Childrens' Mercy: founded in '11, 7476 familys sequenced. 14TB ram and 58 node, 2200-cores.
9:30am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: PDF reports, filed away, and never looked at again. 36% 'continued in error'; 18% 'not aware of gt reports'
9:28am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: 6mos later, 42 given other antiviral also contraindicated. Saw higher viral load, lower CD4 cts
#ampeurope2018 Hoffman: HIV genotype, matched with molecular and pt data. 441 w/resistant GT, 239 had contraindicated Rx !
9:27am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: Pt outcomes: resp to Rx; operational: cost-effective. Ex: '07 https://t.co/FRbLlEGuKa
9:26am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: Success req's data integration, integrate clinical data, molecular data, and financial data.
9:25am May 2nd 2018 via Twitter Web Client
#ampeurope2018 Hoffman: 20K genes, and 3B nucleotides x 2. Devices and EHR; and the individual (apps, search, tweets...) :)
#ampeurope2018 Hoffman: Data explosion: 10^10 neurons, 10^14 connections. 250K-1M unique protiens. 10T microbial cells/indiv
9:24am May 2nd 2018 via Twitter Web Client
