I just saw this press release and thought it worth noting here. I saw the story at Time Magazine yesterday, but I already had a few posts going up them. I find it good that autism research is high enough profile to make the Time list.
PHILADELPHIA, Dec. 11 /PRNewswire-USNewswire/ — Autism research led by scientists at the Children’s Hospital of Philadelphia has been named one of the top ten medical breakthroughs of 2009 by Time Magazine.
On the magazine’s website on Dec. 8, Time cited the largest-ever genetic study of autism spectrum disorders (ASDs), published in April in the journal Nature, by a group led by Hakon Hakonarson, M.D., Ph.D., director of the Center for Applied Genomics at the Children’s Hospital of Philadelphia. That study identified DNA variations that account for as many as 15 percent of all ASD cases. Because the gene region affects how brain cells connect with each other in early childhood, the research significantly advances the understanding of how autism originates.
“We are proud of this research discovery, and are glad to see it receive this recognition,” said Philip R. Johnson, M.D., chief scientific officer at the Children’s Hospital of Philadelphia. “It provides a starting point for translating biological knowledge into future autism treatments.”
The autism gene research from Children’s Hospital, which included two studies in the same issue of Nature, received extensive news coverage, including the CBS Evening News, ABC World News Tonight, BBC, Reuters, the Chicago Tribune, the Philadelphia Inquirer, and other news outlets in the U.K., India, Australia, Germany and China. Hakonarson’s main collaborator was neuroscientist Gerard D. Schellenberg, Ph.D., of the University of Pennsylvania School of Medicine, with other scientists participating from 14 additional centers.
To see Time’s description of new research on autism, click here:
http://www.time.com/time/specials/packages/article/0,28804,1945379_1944376_1944404,00.html
About the Children’s Hospital of Philadelphia: The Children’s Hospital of Philadelphia was founded in 1855 as the nation’s first pediatric hospital. Through its long-standing commitment to providing exceptional patient care, training new generations of pediatric healthcare professionals and pioneering major research initiatives, Children’s Hospital has fostered many discoveries that have benefited children worldwide. Its pediatric research program is among the largest in the country, ranking third in National Institutes of Health funding. In addition, its unique family-centered care and public service programs have brought the 441-bed hospital recognition as a leading advocate for children and adolescents. For more information, visit http://www.chop.edu.
CONTACT: Rachel Salis-Silverman of the Children’s Hospital of Philadelphia, +1-267-426-6063, Salis@email.chop.edu
SOURCE The Children’s Hospital of Philadelphia
Left Brain Right Brain 
That old chestnut “as many as 15%”
Well I have heard the one about the half empty and the half full bottle, but if I found one only 15% full I would be sure that it was 85% empty.
There can be only two justifications for genetic research.
Elimination, either by termination or medication
or idle curiosity.
Even if it is just for ‘idle’ curiosity (a loaded term) it is still justified in the former terms
“It provides a starting point for translating biological knowledge into future autism treatments.”
So just what does make these folk any better than the other bio med merchants, goals are the same, for big pharma to turn a profit.
I would rather the mechanisms of the brain are given more study, than the the rather dubious differences that lead to them.
If I want to alter the performance of my car’s engine, I want to know how it works not who the designers were.
Eventually genetics will cease to be the new black, it will run out of steam once the majority of the studies are discovered to be leading nowhere.
There are deeper mysteries than the human genome, so let that occupy the ‘idle’ curiosity of the scientifically inclined instead.
Whoops spotted an ambiguity in how a reader may interpret intentions in this line:
Even if it is just for ‘idle’ curiosity (a loaded term) it is still justified in the former terms
That should read, it is still justified (by the proponents) in the former terms. I don’t condone that particular justification, it is the one the curious tend to revert to.
This study could also qualify for the ten worst list.
When the study was published the Phildelphia Hospital put on a full-court media blitz hyperexagerating Hakonarnon’s claims. Here is an example of Hakonarnon’s media blitz, this time on Katie Couric’s CBS Evening News.
Nowhere in the interview did Hakonarnon state that the same common genetic variation that was found in 65% of the Autism group is also found in 60% of the control group.
http://www.cbsnews.com/stories/2009/04/28/eveningnews/main4975659.shtml?tag=currentVideoInfo;videoMetaInfo
Hakonarnon did not inform Katie Coric, Sheryl Atkinson or Geraldine Dawson that they have a 60% chance of carrying the ‘autism’ gene.
Well designed studies, especially in autism, control for ethnicity, age, IQ and gender.
http://www.ncbi.nlm.nih.gov/pubmed/19292899?
Common genetic variants can vary be gender:
http://carcin.oxfordjournals.org/cgi/content/abstract/bgp086v1
The Hakonarnon study did not control for a gender bias. The male/female ratio in the autistic cohort was Male=83%, Female = 17%. The male/female ratio in the control group was Male=52%, Female=48%.
The entire difference between the 65% (autism group) and the 60% (control group) could be easily explained by an unrepresentative control group with respect to gender.
RAJ,
I remember quote well this study and your complaints about gender. I also remember that you speculated on and on about that without ever contacting the author. I recall (and have just checked) that when I checked with the author and posted a comment about it, you disappeared without acknowledging that you were speaking from an easily remedied ignorance.
From a personal communication with the author:
RAJ, you arguments about gender bias are incorrect. The fact that you cling to them even after the author’s words were presented demonstrates to this observer a strong bias on your own part.
Actually, CHOP is also doing some interesting work using MRI to track brain activity for recognizing faces in NT vs ASD children. But I guess that part of research was not funded by AS. But you know, Bob Wright, former NBC executive has the inside influence on Time/Warner (who is incidently looking into a TV media partner)…
If you look at their paper
http://www.plosgenetics.org/article/info%3Adoi%2F10.1371%2Fjournal.pgen.1000536#s4
and the last table (Table S5) where they list how many of which type of deletion/duplication there were, there are only 6 markers (out of 361) that did not show up in unaffected controls.
If you look at the fraction of carriers compared to carriers plus affected and average over all markers, the average is 0.47. That is the markers are found 47% of the time the markers were found, they were found in unaffected individuals. This was a study of the extended family of affected individuals, so it is not a surprise that there would be commonality of genetic markers. None of those markers were very specific.
I think this makes a “genetic test” for autism quite impossible or impractical. The question is one of specificity, what level of false-positive and false-negative is acceptable. You can make the false-negative arbitrarily low, but only by increasing the false-positives.
From a personal communication with the author:
…this locus we reported on 5p14.1 contributes to both male and female autism and it makes no difference if we use males or females as controls (we loose some power if we drop out 50% of the females but the locus remains significant; the replication is also performed in trios where this has not impact but the bottom line is that MAF is the same in both sexes of healthy controls
RAJ, you arguments about gender bias are incorrect. The fact that you cling to them even after the author’s words were presented demonstrates to this observer a strong bias on your own part.
Gender bias makes a huge difference. They need to explain why they did not subgroup their findings by gender What is the prevelance of this mutation in autistic males compared to control males as well as autistic females compared to control females. An easy calculation. Their refusal to do so raises significant questions as to why they refused to do so.
They certainly give the appearance of cherry picking their own statistics to best fit the model rather than using a representative control group.
This is not unheard of in science, just look at the ClimateGate scandal, where the researchers at East Anglia discussed how to use data that best fit the model.
Even worse is Hakonarnon’s claim that if this mutation could be removed, 15% of autim would simply disappear.
This study did not identify any gene in the 5P14 region. What they reported was a hot spot in the region between genes.
5P14 common genetic variations present in 60% of the general population have been reported in:
Mental retardation:
http://www.ncbi.nlm.nih.gov/pubmed/18266247?
Obesity:
http://www.ncbi.nlm.nih.gov/pubmed/19288561?
Huntington’s Disease:
http://www.ncbi.nlm.nih.gov/pubmed/18481795?
Ovarion Cancer:
http://www.ncbi.nlm.nih.gov/pubmed/19759843?
Height Variation:
http://www.ncbi.nlm.nih.gov/pubmed/19570815?
Hypertension:
http://www.ncbi.nlm.nih.gov/pubmed/18276622?
Schizophrenia:
http://www.ncbi.nlm.nih.gov/pubmed/17897812?
If we could just remove the 5P14 autism genetic mutation carried by the majority of the worlds population, not only could we eliminate autism, we could also eliminate a significant number of cases of mental retardation, as well as Huntington’s Disease, Obesity, Ovarion Cancer, Hypertension, Height variances, and Schizophrenia.
Where is the specifity to autism?
RAJ,
you are going through the same arguments you made when this first came out. Gender bias is only an issue if it has a difference by gender. It appears–according to the authors own words–that there isn’t a difference.
If you really cared about the answer, you would bring this up with the authors–just like before. If you really just want a weak hammer to criticize an autism/genetics study, you will just continue in the same mode you are in. I’ve done your homework once. I don’t feel like doing it again.
Here is another study published by Hakonarson’s Philadelphia group with autistic genetic data taken from the same AGRE data set that was published the same month as his 5P14 paper:
http://www.ncbi.nlm.nih.gov/pubmed/19557195?
The study implicated whole gene duplications encompassing UBE3A and several other genes in the 15q11-q13 region.
These are the same genetic mutations associated with Angleman Syndrome, a severe neurogenetic disorder ‘characterized by developmental delay, severe intellectual disability, absent speech, exuberant behavior with happy demeanor, motor impairment, and epilepsy’.
http://www.ncbi.nlm.nih.gov/pubmed/9585605?
Is Angelman Syndrome an ‘autistic’ syndrome? It depends on who you ask. Autism researchers looking at an Angelman child see ‘autism’. Families and researchers who specialize in the disorder do not. The Angelman Syndrome foundation which includes families and researchers devoted to Angelman syndrome, on their website state unequivically that Angelman Syndrome is often misdiagnosed as cerebral palsy or autism:
http://www.angelman.org/stay-informed/
What are possible Angelman Syndrome people doing in the AGRE data set? Not hard to figure. The only exclusionary criteria for being accepted into the AGRE data set is the presence of Fragile X. There are cases of Down Syndrome children also present in the AGRE data set.
AGRE appear to be overly inclusive of genetically influenced mental retardation syndromes and the association with ‘autism’ is unclear.The AGRE data set can also be described as not entirely representative of ‘autism’ and as such, any claims made, taken from the AGRE data set should be viewed
with a great deal of scepticism.
Hakonarson’s study with data taken from the AGRE cohort found genetic variants (microdeletions and micro duplications) in the following regions: 15q11-13, 22q11.21, 16p11.2
http://www.nature.com/nature/journal/v459/n7246/full/nature07953.html
All three of the mutations are genetically influenced mental retardation syndrome:
15Q11-13
http://www.ncbi.nlm.nih.gov/pubmed/19748638?
22q11.21
http://www.ncbi.nlm.nih.gov/pubmed/19193630?
16p11.2
http://www.ncbi.nlm.nih.gov/pubmed/19306953?
Autism diagnostic schemes cannot control for mental retardation with ‘autistic type’ symptoms. There is a flaw in the design of the AGRE data sets. The design specifications are families with two or more family members diagnosed with autism, whose exlusionary criteria disqualifies only Fagile X families.
Since the inclusionary criteria is two or more family members, it is clear that the AGRE data set is overly inclusive of genetically influenced mental retardation syndromes.
None of the candidate genes that are present in the AGRE cohort are specific to ‘autism’.
AGRE data cannot be viewed with anything other than complete skepticism
RAJ,
“you are going through the same arguments you made when this first came out. Gender bias is only an issue if it has a difference by gender. It appears—according to the authors own words—that there isn’t a difference.
If you really cared about the answer, you would bring this up with the authors—just like before. If you really just want a weak hammer to criticize an autism/genetics study, you will just continue in the same mode you are in. I’ve done your homework once. I don’t feel like doing it again”.
65% compared to 60% is a meaningless association. My complaint of Hakonarnson is his brazen media blitz at the time of the study where he claimed he had identified the cause of autism in 2/3rds of cases. In none of his media appearances did he state that the same common genetic variant is found in 60% of the general popuilation. He also told a Chicago Tribune science writer that if this autism genetic mutation could be eliminated 15% of autism would simply disappear.
He has already been shown to be morally bankrupt. When he left DeCode genetics in Iceland he was accused by the company of stealing secrets and then plagiarism.
http://www.sciencemag.org/cgi/content/summary/sci;314/5799/580b?maxtoshow=&HITS=10&hits=10&RESULTFORMAT=&fulltext=%28Hakonarson+AND+DeCode+AND+plagiarism%29&searchid=1&FIRSTINDEX=0&resourcetype=HWCIT
It is one thing to tell the media the results of a study, it is quite another when he deliberatley hides the facts and hyperexxagerates the meaning of his study.
http://www.cbsnews.com/video/watch/?id=4975818n&tag=related;photovideo
According to RAJ, every genetic study of autism is flawed because of gender bias, poor phenotypic criteria, faulty or exaggerated risk allele analysis, benign CNVs and moral bankruptcy.
It’s amazing that all these investigators have so consistently fooled their funding sources and publications like Nature, Nature Genetics, PLOS, and Proceedings of.