StarZero All shows Try the podcast clipper →

Jan 2, 2019 · 15m · a16z

a16z Podcast | Revisiting the Gene

Gabriel Ott · 6m spoken Carlos Araya · 3m spoken Jorge Conde · 3m spoken Sonal Chokshi · 33s spoken
0:00 / 0:00
▶ Watch on YouTube →

gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions

This episode of the a16z Podcast explores how human genomics is evolving beyond basic DNA sequencing into actionable clinical diagnostics, highlighting machine learning applications for early cancer detection, genetic variant interpretation, and new healthcare reimbursement models.

How this conversation actually went

Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. The host holds 4% of the talking time here. How this is scored →

The host as informed peer 5.7 Guest teaching 5.3 Guest disagreement 1.3 The host pushing back 3.3
05100:0010:000:37–5:03 · The host as informed peer 5/10 The Challenge and Cost of Genetic Variant Interpretation Jorge establishes a strong technical foundation by referencing genome costs and historical literature before asking targeted questions about variant interpretation. Carlos educates on the massive financial bottleneck in variant interpretation, noting that while sequencing costs $1,000, interpreting novel variants per patient can cost 100-to-1000 times more.5:03–9:28 · The host as informed peer 5/10 Dynamic DNA and Early Disease Detection with Freenome Jorge demonstrates high expertise by synthesizing Gabe's explanation of dynamic DNA into a clear analogy comparing sequence-once models with recurring diagnostic queries like annual dental X-rays. Gabe explains the fundamental biology of dynamic gene expression versus static DNA, clarifying that Freenome measures immune cell turnover in blood.9:28–13:26 · The host as informed peer 7/10 Clinical Misinterpretation Risks and Machine Learning Diagnostics Jorge exhibits high host expertise and pushback by bringing up a real-world Oregon lawsuit regarding genetic misinterpretation and challenging Gabe with historical mammography data showing early detection has not reduced late-stage cancer rates. Gabe responds by revealing mammography's 50 percent false positive rate and detailing how AI feedback loops prevent recurring diagnostic errors.0:37–5:03 · Guest teaching 5/10 The Challenge and Cost of Genetic Variant Interpretation Jorge establishes a strong technical foundation by referencing genome costs and historical literature before asking targeted questions about variant interpretation. Carlos educates on the massive financial bottleneck in variant interpretation, noting that while sequencing costs $1,000, interpreting novel variants per patient can cost 100-to-1000 times more.5:03–9:28 · Guest teaching 5/10 Dynamic DNA and Early Disease Detection with Freenome Jorge demonstrates high expertise by synthesizing Gabe's explanation of dynamic DNA into a clear analogy comparing sequence-once models with recurring diagnostic queries like annual dental X-rays. Gabe explains the fundamental biology of dynamic gene expression versus static DNA, clarifying that Freenome measures immune cell turnover in blood.9:28–13:26 · Guest teaching 6/10 Clinical Misinterpretation Risks and Machine Learning Diagnostics Jorge exhibits high host expertise and pushback by bringing up a real-world Oregon lawsuit regarding genetic misinterpretation and challenging Gabe with historical mammography data showing early detection has not reduced late-stage cancer rates. Gabe responds by revealing mammography's 50 percent false positive rate and detailing how AI feedback loops prevent recurring diagnostic errors.0:37–5:03 · Guest disagreement 1/10 The Challenge and Cost of Genetic Variant Interpretation Jorge establishes a strong technical foundation by referencing genome costs and historical literature before asking targeted questions about variant interpretation. Carlos educates on the massive financial bottleneck in variant interpretation, noting that while sequencing costs $1,000, interpreting novel variants per patient can cost 100-to-1000 times more.5:03–9:28 · Guest disagreement 1/10 Dynamic DNA and Early Disease Detection with Freenome Jorge demonstrates high expertise by synthesizing Gabe's explanation of dynamic DNA into a clear analogy comparing sequence-once models with recurring diagnostic queries like annual dental X-rays. Gabe explains the fundamental biology of dynamic gene expression versus static DNA, clarifying that Freenome measures immune cell turnover in blood.9:28–13:26 · Guest disagreement 2/10 Clinical Misinterpretation Risks and Machine Learning Diagnostics Jorge exhibits high host expertise and pushback by bringing up a real-world Oregon lawsuit regarding genetic misinterpretation and challenging Gabe with historical mammography data showing early detection has not reduced late-stage cancer rates. Gabe responds by revealing mammography's 50 percent false positive rate and detailing how AI feedback loops prevent recurring diagnostic errors.0:37–5:03 · The host pushing back 2/10 The Challenge and Cost of Genetic Variant Interpretation Jorge establishes a strong technical foundation by referencing genome costs and historical literature before asking targeted questions about variant interpretation. Carlos educates on the massive financial bottleneck in variant interpretation, noting that while sequencing costs $1,000, interpreting novel variants per patient can cost 100-to-1000 times more.5:03–9:28 · The host pushing back 2/10 Dynamic DNA and Early Disease Detection with Freenome Jorge demonstrates high expertise by synthesizing Gabe's explanation of dynamic DNA into a clear analogy comparing sequence-once models with recurring diagnostic queries like annual dental X-rays. Gabe explains the fundamental biology of dynamic gene expression versus static DNA, clarifying that Freenome measures immune cell turnover in blood.9:28–13:26 · The host pushing back 6/10 Clinical Misinterpretation Risks and Machine Learning Diagnostics Jorge exhibits high host expertise and pushback by bringing up a real-world Oregon lawsuit regarding genetic misinterpretation and challenging Gabe with historical mammography data showing early detection has not reduced late-stage cancer rates. Gabe responds by revealing mammography's 50 percent false positive rate and detailing how AI feedback loops prevent recurring diagnostic errors.

speaking balance: gold is the host, purple is the guest (3 minute bins)

0:00 · the host 20.2% · guest 79.8%0:00 · the host 20.2% · guest 79.8%3:00 · the host 0% · guest 100%3:00 · the host 0% · guest 100%6:00 · the host 0% · guest 100%6:00 · the host 0% · guest 100%9:00 · the host 0% · guest 100%9:00 · the host 0% · guest 100%12:00 · the host 0% · guest 100%12:00 · the host 0% · guest 100%15:00 · the host 0% · guest 100%15:00 · the host 0% · guest 100%
Sharpest disagreement ▶ 11:21 Dismissing Mammography Accuracy

Gabe forcefully reframes traditional screening technology by pointing out that mammography has a 50 percent false positive rate, claiming a patient is better off flipping a coin.

 Hardest push from the host ▶ 10:50 Mammography Over-Diagnosis Challenge

Jorge refuses to accept early detection as inherently beneficial, pressing Gabe with 30-40 years of mammography data showing early screening failed to lower late-stage cancer rates.

 Biggest teaching moment ▶ 5:12 Educating on Dynamic vs. Static DNA

Gabe educates the host on how consumer tests like 23andMe examine under one percent of static DNA, whereas dynamic DNA expression determines real-time cellular health.

 The host holds their own ▶ 9:28 Citing Oregon Misinterpretation Lawsuit

Jorge demonstrates deep industry knowledge by raising a specific Oregon lawsuit where misinterpreting a genetic variant of unknown significance led to an unnecessary medical procedure.
the scores for every segment, with the reasoning behind each
ChapterTopicThe host as informed peerGuest teachingGuest disagreementThe host pushing backWhy
The Challenge and Cost of Genetic Variant Interpretation 5512 Jorge establishes a strong technical foundation by referencing genome costs and historical literature before asking targeted questions about variant interpretation. Carlos educates on the massive financial bottleneck in variant interpretation, noting that while sequencing costs $1,000, interpreting novel variants per patient can cost 100-to-1000 times more.
Dynamic DNA and Early Disease Detection with Freenome 5512 Jorge demonstrates high expertise by synthesizing Gabe's explanation of dynamic DNA into a clear analogy comparing sequence-once models with recurring diagnostic queries like annual dental X-rays. Gabe explains the fundamental biology of dynamic gene expression versus static DNA, clarifying that Freenome measures immune cell turnover in blood.
Clinical Misinterpretation Risks and Machine Learning Diagnostics 7626 Jorge exhibits high host expertise and pushback by bringing up a real-world Oregon lawsuit regarding genetic misinterpretation and challenging Gabe with historical mammography data showing early detection has not reduced late-stage cancer rates. Gabe responds by revealing mammography's 50 percent false positive rate and detailing how AI feedback loops prevent recurring diagnostic errors.

Statements from this episode (10)

Assertion Contradicted
Each sequenced genome contains about 100 novel variants in disease genes
“Although there are three million variants identified, There will be roughly a hundred variants that are novel variants in disease-associated genes.”
Carlos Araya Jan 2, 2019 ▶ 2:48
Assertion Supported
Interpreting novel genetic variants costs 100x to 1000x more than sequencing
“And interpreting each one of those under the current clinical practices costs 50 to a hundred dollars. So we're talking about a hundred to thousand-fold increase in the cost of interpretation relative to the cost of data acquisition.”
Carlos Araya Jan 2, 2019 ▶ 2:59
Assertion Supported
Only 0.6% of disease-associated gene mutations have clinical interpretations
“If we look across all of the disease-associated genes that we know today, we only have clinical interpretations for roughly . Six percent of the possible mutations in them.”
Carlos Araya Jan 2, 2019 ▶ 3:52
Disclosure
Freenome detects dynamic cell-free DNA from circulating immune cells
“What we're detecting is DNA fragments that are actually coming from the immune cells that are turning over in your body.”
Gabriel Ott Jan 2, 2019 ▶ 8:39
Prediction Not checkable as stated
Ott: Dynamic DNA testing will enable detection of any disease with immune changes
“The underlying biology should theoretically enable us to detect any diseases where there is an immune change.”
Gabriel Ott Jan 2, 2019 ▶ 9:18
Assertion Partly supported
Ott: Mammography Screenings Have a 50% False Positive Rate
“Breast cancer, and specifically mammography as a screening method, has a false positive rate of 50%.”
Gabriel Ott Jan 2, 2019 ▶ 11:30
Insight
AI diagnostics can continuously improve accuracy post-launch by learning from results
“Because we can now make a diagnostic that's fundamentally AI-based, even after we launch the test into the market, we can actually work with our partners to get results of the test that we sell back so that we can teach the artificial intelligence that it made…”
Gabriel Ott Jan 2, 2019 ▶ 12:20
Assertion Supported
Ott: 35 Million Americans Missed Colorectal Cancer Screenings Last Year
“Last year, thirty-five million people should have gotten screened for colorectal cancer in the United States alone and didn't.”
Gabriel Ott Jan 2, 2019 ▶ 12:58
Assertion Supported
Ott: Only 20% of new diagnostic tests receive full reimbursement
“If you're looking at the average statistics, generally when you're launching a diagnostic test, only about 20% of the tests that you sell actually get fully reimbursed.”
Gabriel Ott Jan 2, 2019 ▶ 13:39
Prediction Not checkable as stated
Diagnostic tests will increasingly leverage non-traditional payment models like life insurance
“I think we're going to see leveraging of these new models much more, but it's still in the early days.”
Gabriel Ott Jan 2, 2019 ▶ 14:44
Made with StarZero

Turn any episode into a week of clips.

This entire site, over 1,000 episodes transcribed, diarized, checked and made playable, runs on the StarZero media pipeline. Drop in your own episode and the podcast clipper finds the moments worth sharing, cuts them, captions them, and reframes them for every feed.

Try the podcast clipper Learn more