Isobel Beasley
@ijbeasley.bsky.social
Biological and Medical Informatics PhD student @UCSF by day, hobby collector and lemur lover by night. Working towards a world where everyone is included in, and benefits from, genetics research.
🏳️🌈 she/they
isobelbeasley.com
🏳️🌈 she/they
isobelbeasley.com
Reposted by Isobel Beasley
Here is a word cloud of the words that were missing from the noncompetitive renewal awards but had been present in the parent award.
3/4
3/4
September 9, 2025 at 11:22 PM
Here is a word cloud of the words that were missing from the noncompetitive renewal awards but had been present in the parent award.
3/4
3/4
In general, thresholds provide a clear and consistent answer to 'how do I treat this patient in front of me' - making them very appealing to busy clinicians.
August 18, 2025 at 2:53 PM
In general, thresholds provide a clear and consistent answer to 'how do I treat this patient in front of me' - making them very appealing to busy clinicians.
In terms of why we still use binary prs - I think it's reflective of current medical practice. Using thresholds to get yes/no assessments of risk / disease status is ubiquitous even though underlying continous measurements often provide more valuable information (e.g. high blood pressure cut-offs).
August 18, 2025 at 2:53 PM
In terms of why we still use binary prs - I think it's reflective of current medical practice. Using thresholds to get yes/no assessments of risk / disease status is ubiquitous even though underlying continous measurements often provide more valuable information (e.g. high blood pressure cut-offs).
2) where it does hold, reducing a continuous measure of risk to a binary yes/no option reduces predictive power (and requires choosing a good threshold - which can be very hard)
August 18, 2025 at 2:53 PM
2) where it does hold, reducing a continuous measure of risk to a binary yes/no option reduces predictive power (and requires choosing a good threshold - which can be very hard)
My understanding is the main issues are:
1) liability threshold model doesn't always hold (e.g. non-normality of risk, environmentally dependent effects, sex + age etc. differences in risk - this paper goes a bit into the first two: pmc.ncbi.nlm.nih.gov/articles/PMC...)
1) liability threshold model doesn't always hold (e.g. non-normality of risk, environmentally dependent effects, sex + age etc. differences in risk - this paper goes a bit into the first two: pmc.ncbi.nlm.nih.gov/articles/PMC...)
How meaningful are heritability estimates of liability?
It is commonly acknowledged that estimates of heritability from classical twin studies have many potential shortcomings. Despite this, in the post-GWAS era, these heritability estimates have come to b...
pmc.ncbi.nlm.nih.gov
August 18, 2025 at 2:53 PM
My understanding is the main issues are:
1) liability threshold model doesn't always hold (e.g. non-normality of risk, environmentally dependent effects, sex + age etc. differences in risk - this paper goes a bit into the first two: pmc.ncbi.nlm.nih.gov/articles/PMC...)
1) liability threshold model doesn't always hold (e.g. non-normality of risk, environmentally dependent effects, sex + age etc. differences in risk - this paper goes a bit into the first two: pmc.ncbi.nlm.nih.gov/articles/PMC...)
Reposted by Isobel Beasley
Germany used to lead the world in science. The language of science was German—until the Nazis came to power and chased out, imprisoned, or killed all the non-Nazi scientists. Fascists break good things, and it is happening here. (2/2)
July 28, 2025 at 5:54 PM
Germany used to lead the world in science. The language of science was German—until the Nazis came to power and chased out, imprisoned, or killed all the non-Nazi scientists. Fascists break good things, and it is happening here. (2/2)