7 ms·

Sperm DNA methylation epimutation biomarker for paternal offspring autism

75 points by johntfella 5y ago

johntfella 5y ago
Full title: Sperm DNA methylation epimutation biomarker for paternal offspring autism susceptibility
dvfjsdhgfv 5y ago
Does anyone know of a similar study but involving mothers instead of fathers?
- nightfly 5y ago
  It's a lot more destructive to harvest eggs than sperm.
  - aminozuur 5y ago
    Fertility clinics harvest eggs from women every day. Some of those eggs could be donated for research. Example: Clinics harvests 12 eggs from a layd, 1 egg needed for IVF, IVF is succesful (baby born), lady is now happy to donate the remaining 11 eggs to science.
    - jschwartzi 5y ago
      Some nitpicks: * IVF involves retrieval and then fertilization of eggs. There’s not usually leftover unfertilized eggs unless the woman is banking her eggs. * The success rates my partner and I have been quoted mean that we want to keep as many embryos as possible in case we need to try again. I think it would be really unethical to ask a woman to give up any good eggs or euploid embryos as they can be hard to get in some cases. The shots a woman has to give herself daily to stimulate follicle production are no joke. That would be a lot to put someone through purely for research. It’s not as simple as that. And it’s wishful thinking to expect any woman to do that.
- ultimape 5y ago
  I haven't found one specifically on autism. Most of the mothers' risk in offspring seems to be gestational factors during brain development stages, and later on impacts on milk. This gets close "Epigenetics in Families: Covariance between Mother and Child Methylation Patterns" https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7913850/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7913850/
skissane 5y ago
Two samples each with n = 13. ASD is widely acknowledged to be a highly heterogeneous condition which is likely to have multiple independent causes, and which has unclear boundaries with other diagnoses (especially ADHD) So a study which effectively treats ASD as a single undifferentiated condition, and has rather small sample sizes too, I think should be treated with a great deal of scepticism.
- 2T1Qka0rEiPr 5y ago
  Yes, I read down to this point and stopped. What significance can you really draw from a sample size of 26? (Serious question to academics out there)
  - renewiltord 5y ago
    To answer your serious question, n=26 is more than sufficient depending on effect size and sampling method. Think about it, if you have 26 people randomly sampled and you give half a pill and the other half a placebo and the first half dies moments after ingestion, how confident are you that the pill is poisonous? If the effect size is small, you may fail to detect it with a small population. If the sampling is biased, then you’re going to have a problem even if you have a massive population studied.
  - tpoacher 5y ago
    The "small sample size" has been drilled to most people as a knee jerk red flag indicating a statistically unreliable conclusion, but this is not always the case. Depending on the desired outcome, a small sample size may actually increase the reliability of the result. This seems like one of those cases. Think of it like this. The aim is to show a statistically significant difference. With a large enough sample, you will eventually show significance (hence Fisher's famous anecdote of "get more data"), but the effect size may be trivial. Whereas, with a small sample, only non-trivial effect sizes will achieve significance, and therefore achieving significance with such a small sample tells you something about the nontrivial extent of the effect size. So you could argue that the extent of methylation change was so large, that it achieved significance in a sample even as small as n=13!
h0l0cube 5y ago
For some this opinion might be contentious/taboo, but mild autism is prevalent across much of the skilled labor pool in STEM - and I count myself within that statistic. So it'll be interesting in a couple of generations what might happen to this labor pool if the potential to screen for the broad autism phenotype affects who is born and/or affects sexual selection.
- bozzcl 5y ago
  Is it contentious? I thought it was pretty accepted. Pretty obvious too, even just looking at the people I know.
  - mywittyname 5y ago
    It hasn't really been studied or formalized in any way (at least that I know of). But I'm confident that it will be found to be the case if/when it is studied.
    - h0l0cube 5y ago
      Yeah this is what I meant. I think that 'taboo' is also a big part of it. I think there are many high functioning autistic people that really don't want the label/perception, but also seem to be burning themselves out by constantly masking (to maybe little benefit). Seems like they could be happier if they accepted who they were
- 1-6 5y ago
  You say statistic. Is there a document regarding this?
  - skissane 5y ago
    https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0141229 https://journals.plos.org/plosone/article?id=10.1371/journal... Autism Quotient (AQ) scores are generally accepted as a measure of Broad Autism Phenotype (BAP), which is subclinical ASD. That study produces the (entirely expected) conclusion that BAP is more common in men than women, and more common in STEM occupations than non-STEM occupations. I'm not aware of any studies linking software engineering specifically to BAP, as opposed to STEM occupations in general. I think the evidence on that point is (thus far) based on anecdote and clinical experience.
pvaldes 5y ago
Trying to reconstruct the title Found a mutation in spermatozoa that can be used as a flag for higher probability of having autistic children. Correct?
- Accujack 5y ago
  More like "Found: an epigenetic change in sperm that seems to correlate with genes we know are involved in autism".
- tomp 5y ago
  Methilation isn't a mutation. It's epigenetics - (potentially) persistent / heritable changes in phenotype (expressed genes) outside of genetic code (DNA). It's used for many things, e.g. turning on/off specific genes in somatic cells, repairing DNA (when DNA is copied, the old strand is methilated; so you know how to fix any differences between old and new strand), and more recently, for measuring aging (Horvath epigenetic clock).
  - pvaldes 5y ago
    We can substitute a mutation as a synonym for modification here then. It doesn't matter and would be still more clear. We could say that the string of letters is not modified but the letters in the DNA string are "in cursive" now, so when we read the text we understand a different thing.
TeeMassive 5y ago
> Methods and results > Sperm samples were obtained from fathers that have children with or without autism, and the sperm then assessed for alterations in DNA methylation. A genome-wide analysis (> 90%) for differential DNA methylation regions (DMRs) was used to identify DMRs in the sperm of fathers (n = 13) with autistic children in comparison with those (n = 13) without ASD children. The 805 DMR genomic features such as chromosomal location, CpG density and length of the DMRs were characterized. Genes associated with the DMRs were identified and found to be linked to previously known ASD genes, as well as other neurobiology-related genes. The potential sperm DMR biomarkers/diagnostic was validated with blinded test sets (n = 8–10) of individuals with an approximately 90% accuracy. This has to be a preliminary study right? I mean n=13...
fallingfrog 5y ago
“One of the control samples, IVI 14, had a very high sperm count of 396.62 million that was outside two standard deviations of the mean (2 ± SD), so the analysis was redone without this sample.” Yeah having chuck norris in your study is going to skew any set of results.
seanhandley 5y ago
> Autism spectrum disorder (ASD) has increased over tenfold over the past several decades That should be "Diagnosis of ASD has increased tenfold..." because of a heap of factors including the medical criteria of diagnosis changing, a historical underdiagnosis of autism in girls and women, and improved public knowledge of the condition leading to more assessments and diagnoses. EDIT: Towards the end, it says > The frequency of autism in the population has dramatically increased over tenfold the past several decades. This increase appears to be due in part to increased diagnosis efficiency from 1975 to the early 2000s, as well as greater public awareness of the disease [3]. The more recent increase in the last couple of decades suggests environmental factors, and exposures also have a critical role in autism prevalence.
- ghostpepper 5y ago
  So are you saying our ability to detect it has not actually improved in the past several decades?
  - seanhandley 5y ago
    I'm saying an increase in diagnoses recently is in large part due to a historical underdiagnosis rather than an increasing frequency of children born with ASD. However, the paper hints at environmental factors being responsible for a proportion of the increase (but does not cite evidence to prove it).
    - brutus1213 5y ago
      Paper seems to refer to environment factors and paternal DNA. Is this established or just a hypothesis (the paternal part especially)? It is kind of sad given the advances in sequencing and state of computing power that we cannot reliably diagnose this stuff with sequencing. I thought the promise of 23andme, and ilk was to gain large amounts of data and make progress on these problems (NIPT seems like a success though).

75 points by johntfella 5y ago

johntfella 5y ago

Full title: Sperm DNA methylation epimutation biomarker for paternal offspring autism susceptibility

dvfjsdhgfv 5y ago

Does anyone know of a similar study but involving mothers instead of fathers?

nightfly 5y ago
It's a lot more destructive to harvest eggs than sperm.
- aminozuur 5y ago
  Fertility clinics harvest eggs from women every day. Some of those eggs could be donated for research. Example: Clinics harvests 12 eggs from a layd, 1 egg needed for IVF, IVF is succesful (baby born), lady is now happy to donate the remaining 11 eggs to science.
  - jschwartzi 5y ago
    Some nitpicks: * IVF involves retrieval and then fertilization of eggs. There’s not usually leftover unfertilized eggs unless the woman is banking her eggs. * The success rates my partner and I have been quoted mean that we want to keep as many embryos as possible in case we need to try again. I think it would be really unethical to ask a woman to give up any good eggs or euploid embryos as they can be hard to get in some cases. The shots a woman has to give herself daily to stimulate follicle production are no joke. That would be a lot to put someone through purely for research. It’s not as simple as that. And it’s wishful thinking to expect any woman to do that.
ultimape 5y ago
I haven't found one specifically on autism. Most of the mothers' risk in offspring seems to be gestational factors during brain development stages, and later on impacts on milk. This gets close "Epigenetics in Families: Covariance between Mother and Child Methylation Patterns" https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7913850/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7913850/

skissane 5y ago

Two samples each with n = 13. ASD is widely acknowledged to be a highly heterogeneous condition which is likely to have multiple independent causes, and which has unclear boundaries with other diagnoses (especially ADHD) So a study which effectively treats ASD as a single undifferentiated condition, and has rather small sample sizes too, I think should be treated with a great deal of scepticism.

2T1Qka0rEiPr 5y ago
Yes, I read down to this point and stopped. What significance can you really draw from a sample size of 26? (Serious question to academics out there)
- renewiltord 5y ago
  To answer your serious question, n=26 is more than sufficient depending on effect size and sampling method. Think about it, if you have 26 people randomly sampled and you give half a pill and the other half a placebo and the first half dies moments after ingestion, how confident are you that the pill is poisonous? If the effect size is small, you may fail to detect it with a small population. If the sampling is biased, then you’re going to have a problem even if you have a massive population studied.
- tpoacher 5y ago
  The "small sample size" has been drilled to most people as a knee jerk red flag indicating a statistically unreliable conclusion, but this is not always the case. Depending on the desired outcome, a small sample size may actually increase the reliability of the result. This seems like one of those cases. Think of it like this. The aim is to show a statistically significant difference. With a large enough sample, you will eventually show significance (hence Fisher's famous anecdote of "get more data"), but the effect size may be trivial. Whereas, with a small sample, only non-trivial effect sizes will achieve significance, and therefore achieving significance with such a small sample tells you something about the nontrivial extent of the effect size. So you could argue that the extent of methylation change was so large, that it achieved significance in a sample even as small as n=13!

h0l0cube 5y ago

For some this opinion might be contentious/taboo, but mild autism is prevalent across much of the skilled labor pool in STEM - and I count myself within that statistic. So it'll be interesting in a couple of generations what might happen to this labor pool if the potential to screen for the broad autism phenotype affects who is born and/or affects sexual selection.

bozzcl 5y ago
Is it contentious? I thought it was pretty accepted. Pretty obvious too, even just looking at the people I know.
- mywittyname 5y ago
  It hasn't really been studied or formalized in any way (at least that I know of). But I'm confident that it will be found to be the case if/when it is studied.
  - h0l0cube 5y ago
    Yeah this is what I meant. I think that 'taboo' is also a big part of it. I think there are many high functioning autistic people that really don't want the label/perception, but also seem to be burning themselves out by constantly masking (to maybe little benefit). Seems like they could be happier if they accepted who they were
1-6 5y ago
You say statistic. Is there a document regarding this?
- skissane 5y ago
  https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0141229 https://journals.plos.org/plosone/article?id=10.1371/journal... Autism Quotient (AQ) scores are generally accepted as a measure of Broad Autism Phenotype (BAP), which is subclinical ASD. That study produces the (entirely expected) conclusion that BAP is more common in men than women, and more common in STEM occupations than non-STEM occupations. I'm not aware of any studies linking software engineering specifically to BAP, as opposed to STEM occupations in general. I think the evidence on that point is (thus far) based on anecdote and clinical experience.

pvaldes 5y ago

Trying to reconstruct the title Found a mutation in spermatozoa that can be used as a flag for higher probability of having autistic children. Correct?

Accujack 5y ago
More like "Found: an epigenetic change in sperm that seems to correlate with genes we know are involved in autism".
tomp 5y ago
Methilation isn't a mutation. It's epigenetics - (potentially) persistent / heritable changes in phenotype (expressed genes) outside of genetic code (DNA). It's used for many things, e.g. turning on/off specific genes in somatic cells, repairing DNA (when DNA is copied, the old strand is methilated; so you know how to fix any differences between old and new strand), and more recently, for measuring aging (Horvath epigenetic clock).
- pvaldes 5y ago
  We can substitute a mutation as a synonym for modification here then. It doesn't matter and would be still more clear. We could say that the string of letters is not modified but the letters in the DNA string are "in cursive" now, so when we read the text we understand a different thing.

TeeMassive 5y ago

> Methods and results > Sperm samples were obtained from fathers that have children with or without autism, and the sperm then assessed for alterations in DNA methylation. A genome-wide analysis (> 90%) for differential DNA methylation regions (DMRs) was used to identify DMRs in the sperm of fathers (n = 13) with autistic children in comparison with those (n = 13) without ASD children. The 805 DMR genomic features such as chromosomal location, CpG density and length of the DMRs were characterized. Genes associated with the DMRs were identified and found to be linked to previously known ASD genes, as well as other neurobiology-related genes. The potential sperm DMR biomarkers/diagnostic was validated with blinded test sets (n = 8–10) of individuals with an approximately 90% accuracy. This has to be a preliminary study right? I mean n=13...

fallingfrog 5y ago

“One of the control samples, IVI 14, had a very high sperm count of 396.62 million that was outside two standard deviations of the mean (2 ± SD), so the analysis was redone without this sample.” Yeah having chuck norris in your study is going to skew any set of results.

seanhandley 5y ago

> Autism spectrum disorder (ASD) has increased over tenfold over the past several decades That should be "Diagnosis of ASD has increased tenfold..." because of a heap of factors including the medical criteria of diagnosis changing, a historical underdiagnosis of autism in girls and women, and improved public knowledge of the condition leading to more assessments and diagnoses. EDIT: Towards the end, it says > The frequency of autism in the population has dramatically increased over tenfold the past several decades. This increase appears to be due in part to increased diagnosis efficiency from 1975 to the early 2000s, as well as greater public awareness of the disease [3]. The more recent increase in the last couple of decades suggests environmental factors, and exposures also have a critical role in autism prevalence.

ghostpepper 5y ago
So are you saying our ability to detect it has not actually improved in the past several decades?
- seanhandley 5y ago
  I'm saying an increase in diagnoses recently is in large part due to a historical underdiagnosis rather than an increasing frequency of children born with ASD. However, the paper hints at environmental factors being responsible for a proportion of the increase (but does not cite evidence to prove it).
  - brutus1213 5y ago
    Paper seems to refer to environment factors and paternal DNA. Is this established or just a hypothesis (the paternal part especially)? It is kind of sad given the advances in sequencing and state of computing power that we cannot reliably diagnose this stuff with sequencing. I thought the promise of 23andme, and ilk was to gain large amounts of data and make progress on these problems (NIPT seems like a success though).