Have questions? Visit https://www.reddit.com/r/SNPedia


From SNPedia

Alternative Population Sources[edit]


? (A;A) (A;T) (T;T) 28


? (A;A) (A;T) (T;T)

Alternative Population Documentation[edit]


KRIBB_YJKIM email: yjkim8@kribb.re.kr fax: 82 42 879 8119 lab: Functional Genomics Research Center name: Young J. Kim address: 52 Eoeun-dong, Yuseong-gu, Daejeon 305-333, Korea phone: 82 42 879 8118 institution: Korea Research Institute of Bioscience and Biotechnology http://www.kribb.re.kr/eng/ http://bib.oxfordjournals.org/content/9/3/189.abstract


Please respond here. John Lloyd Scharf 20:31, 2 November 2011 (UTC)

Currently the preferred source of genotype frequencies is from http://hapmap.ncbi.nlm.nih.gov/ HapMap Release 28 (Phase 1, 2 and 3) for only QC+ SNPs. Currently the newest HapMap data in dbSNP is release 27, however some SNPs have even older HapMap data, or have multiple versions of HapMap data, some of which may be QC-. Due to the variety of problems with the HapMap data in dbSNP, most SNPedia pages are updated directly from the current HapMap data release to ensure the best and most current data. Please note that you should include the "HapMapRevision" field ONLY if you know for certain which release the data comes from. Anything without this field is assumed to be possibly suspect. As of yet, non-HapMap genotype frequencies are not used, though if there is a source with good coverage and accuracy, it can of course be considered. Jlick 05:46, 3 November 2011 (UTC)

  • I documented the sources of all of those as I indicated above. When you go from 12 sources to 6 sources, why not substitute these alternative populations to maintain better resolution and perspective as a whole? Why would their coverage and accuracy be suspect if they are published by the NCIB?
  • What makes the HapMap not be possibly suspect? The CEU data is highly suspect for a number of reasons, which I have mentioned before. You need to come up with a far better reasoning than that.
  • It looks a great deal like there is an American-European bias, unless you can come up with more justifiable and precise criteria for inclusion or exclusion. It should be enough that the NCBI/NIH have included them.
  • The HapMap Project divided up the chromosomes into teams without overlap to challenge each other's assumptions. I certainly do not believe Sanger Institute should be the sole source on SNPs from Chromosome 6. Help me understand why you want to limit your sources to a single institution? John Lloyd Scharf 17:50, 3 November 2011 (UTC)

OK, you are in charge of updating the population templates from now in. Jlick 12:09, 4 November 2011 (UTC)

Perhaps... John Lloyd Scharf 17:29, 4 November 2011 (UTC)

JLick's bots do an amazing amount of the work around here, and do it better than my bots which came before them. When you or your bots manage to do a better job, their edits are most welcome. Until then, the tone of your criticisms is unwelcome. --- cariaso 19:14, 5 November 2011 (UTC)

Is that an invitation to review the bots too? Let's stick to the issues rather than getting personal. John Lloyd Scharf 00:19, 6 November 2011 (UTC)

Reviewing their edits is always welcome. Reviewing their code is up to the author deciding to make their code public (or privately) available. SNPediaBot is not public. I don't believe JLickBot is either. However JLickBot and SNPediaBot have learned to correct or respect each other's edits, and presumably a JohnLloydScharfBot could do the same. In short, if you've got a way to make things better we'd all like to see it. Otherwise, we're doing the best we can. --- cariaso 04:31, 6 November 2011 (UTC)
I do not have a bot so I cannot depend on one. I am not a software tech. I have had not education since Fortran IV and Basic. I am doing the best I can as well. If you have a problem with what I am doing, show me how to improve it rather than telling me you do not "appreciate" it. I can take real criticism. Adjectives are not "appreciated." I do not have a certain way of making things better other than giving MORE information. If that takes a little growth, then I am a patient man. I did not stop at 24, 34, 44, or 54. I do not expect to stop at 64. We all have biases. We are not one of these teams of HapMap that can afford to indulge them. If it were me picking groups to look at, I would have chosen those of Dagestan where 80+% are J1 and the Altai of Central Asia, but that would be biased. However, do not assume when I am questioning something that it has anything to do with "respect." "Sometimes it is not all about you."John Lloyd Scharf 16:30, 6 November 2011 (UTC)

Please remember to remove the "HapMapRevision" line from the template when you make edits using older releases of the data. Jlick 11:18, 6 November 2011 (UTC)

I keep forgetting that, but I do not know where to look for the release number. John Lloyd Scharf 16:12, 6 November 2011 (UTC)

Maybe you should do some more investigation to figure this out before making further updates with old data. Or at least remove the line if you can't be sure. Jlick 18:24, 6 November 2011 (UTC)


  • This group seems to have more donors than HapMap, but no divisions.
  • I am not much on pathology, but if you go that way this sounds good
"The group includes some of the largest well-phenotyped populations in the United States, representing more than 200,000 individuals altogether from the:
Women's Health Initiative (WHI)
Framingham Heart Study (FHS)
Jackson Heart Study (JHS)
Multi-Ethnic Study of Atherosclerosis (MESA)
Atherosclerosis Risk in Communities (ARIC)
Coronary Artery Risk Development in Young Adults (CARDIA)
Cardiovascular Health Study (CHS)
Genomic Research on Asthma in the African Diaspora (GRAAD)
Lung Health Study (LHS)
Pulmonary Arterial Hypertension (PAH) population
Acute Lung Injury (ALI) cohort
Cystic Fibrosis (CF) cohort
  • For questions, please contact Dana Seaman at dseaman@uw.edu