PDA

View Full Version : New ISOGG Y-tree leadership Search



Cofgene
01-24-2016, 10:59 PM
In view of the censorship of Y-tree related discussions which occurred today on the Yahoo ISOGG discussion list - this is the 2nd time scientific discussions have been squashed - I am looking for individuals or a group who would consider and be able to serve as an interim y-tree coordinator. The immediate transition will be a thankless one but the outcome is expected to provide a tree with improved operational processes which improve the quality of the y-tree content and set the stage for significant automation of variant identification and listing.

Ideally the candidate individual(s) should have an advanced degree in the sciences or medical profession. Project management experience would be a bonus for driving.

I am starting this discussion to gather what type of qualifications and organizational structure should be utilized for ISOGG y-tree activities.

This is an open call for an immediate change.

AOK
01-24-2016, 11:51 PM
I was involved with design of the original one in the beginning. I left ISOGG because of <probably doesn't matter>. As we all know, the old tree format we developed is outdated since the flood of SNPs from full Y tests.

I would be willing to give assistance/input if you want it. But not on Yahoo or Google or Facebook. A standard forum like this along with IRC, IM, email list, etc (the old stand-bys) are my preferred tools. I mostly want to see the tree presented in a manageable, useable format, and kept as current as possible.

B.S. & M.S. in Biology (genetics was not the same back then and was not my major), former DNA project organizer/manager (non-FTDNA) and tested with multiple companies, html coding and some (rusty) programming experience, currently working full time (can't check personal email etc. at work).

VinceT
01-25-2016, 03:32 AM
Cofgene:

Please do not forget that "ISOGG" is nothing more than a service mark owned and managed by Mrs. Borges, loosely applied to her personal network of friends and associates. It is not a legally registered corporation, NPO or professional association. It does not have governing bylaws, assets, or an elected board of officers. "ISOGG" cannot and will never be anything other or more than what Mrs. Borges herself will allow it to be.

The Y-tree you speak of, and the Y-tree you hope for, must be hosted and managed independently of "ISOGG". It's as simple, and as challenging, as that.

AOK
01-25-2016, 06:33 AM
And that''s a good idea. The organization is irrelevant to me and most other users. It isn't the same as it was 10 years ago. I'm not even sure we need one tree. We could have links to the parts that are best handled separately.

I have already quit using the ISOGG tree so, in the absence of a good replacement, I will continue to find the best alternative or make my own to suit nyself.

paulgill
01-25-2016, 08:19 AM
And that''s a good idea. The organization is irrelevant to me and most other users. It isn't the same as it was 10 years ago. I'm not even sure we need one tree. We could have links to the parts that are best handled separately.

I have already quit using the ISOGG tree so, in the absence of a good replacement, I will continue to find the best alternative or make my own to suit nyself.

Are you not using this one http://www.yfull.com/tree/

Cofgene
01-25-2016, 12:21 PM
Are you not using this one http://www.yfull.com/tree/

That is not an independent tree. Like FTDNA's internal tree it is dependent upon a subset of the data present in the community. Individuals should not have to pay $$ to have their haplogroup identified and placed on a globally referenced resource.

Cofgene
01-25-2016, 12:22 PM
Cofgene:

Please do not forget that "ISOGG" is nothing more than a service mark owned and managed by Mrs. Borges, loosely applied to her personal network of friends and associates. It is not a legally registered corporation, NPO or professional association. It does not have governing bylaws, assets, or an elected board of officers. "ISOGG" cannot and will never be anything other or more than what Mrs. Borges herself will allow it to be.

The Y-tree you speak of, and the Y-tree you hope for, must be hosted and managed independently of "ISOGG". It's as simple, and as challenging, as that.


I know. :P But we do need to see what type of talent is available to facilitate improved operations of the successor.

Michał
01-25-2016, 03:20 PM
That is not an independent tree. Like FTDNA's internal tree it is dependent upon a subset of the data present in the community. Individuals should not have to pay $$ to have their haplogroup identified and placed on a globally referenced resource.
AFAIK, you don't need to pay to have your haplogroup identified and placed on the YFull tree. You will only need to pay if you want to have access to the more detailed results of the YFull analysis (like the exact result for any SNP or nucleotide position tested, Y-DNA STRs, mtDNA, etc.).

GTC
01-25-2016, 03:57 PM
ISOGG's simple HTML based tree cannot possibly be expected to cope with the tsunami of SNPs flowing from NGS techniques.

I gather that there was some talk a year or more ago within ISOGG about converting it to a model with a database at its center, but nothing came of that.

10 years down the track I think the task is now well beyond the well-meaning ISOGG volunteers and I would prefer any such tree to be managed by professionals, as long as they were independent of the testing companies.

lgmayka
01-25-2016, 04:02 PM
AFAIK, you don't need to pay to have your haplogroup identified and placed on the YFull tree. You will only need to pay if you want to have access to the more detailed results of the YFull analysis (like the exact result for any SNP or nucleotide position tested, Y-DNA STRs, mtDNA, etc.).
This is unclear. Initially, YFull's policy was exactly as you describe. But YFull's web site now has a clause which sounds more like a true service contract (http://www.yfull.com/static/offer-contract.html). In practice, my observation is that if a customer forgets to pay his $49 fee, YFull not only keeps his personal results hidden but also (eventually) removes his entry from the haplotree. Even then, however, YFull's haplotree retains any new tree levels or TMRCAs added during the analysis.

JamesKane
01-25-2016, 04:04 PM
AFAIK, you don't need to pay to have your haplogroup identified and placed on the YFull tree. You will only need to pay if you want to have access to the more detailed results of the YFull analysis (like the exact result for any SNP or nucleotide position tested, Y-DNA STRs, mtDNA, etc.).

I believe YFULL removes the data when their analysis has not been paid for in a timely manner after completion.

GoldenHind
01-25-2016, 07:32 PM
Please do not forget that "ISOGG" is nothing more than a service mark owned and managed by Mrs. Borges, loosely applied to her personal network of friends and associates. It is not a legally registered corporation, NPO or professional association. It does not have governing bylaws, assets, or an elected board of officers. "ISOGG" cannot and will never be anything other or more than what Mrs. Borges herself will allow it to be.

The Y-tree you speak of, and the Y-tree you hope for, must be hosted and managed independently of "ISOGG". It's as simple, and as challenging, as that.

Thanks, I was not aware of that. What I have noticed is that the position for P312 has been vacant for a considerable period.

Unfortunately it appears that when aDNA is analyzed, the ISOGG tree (usually the even more out of date one from 2013) is the basis for their analysis.

Heber
01-25-2016, 09:00 PM
Most phylogenetic trees are the result of both human and computer analytics.
Examples which come to mind are YFull, The Big Tree, The Morley YTree, Fudan University CVTree etc.
The SNP Tsunami caused by DTC NGS testing is causing a data deluge which is overwhelming the existing manual processes and naming conventions despite the heroic efforts of a small group of citizen scientists.

The industry in which I work (telecommunications) solved this problem many years ago by delegating the responsibility of standards to vendor neutral industry associations such as ITU, GSMA, TMForum etc. The competing operators, vendors and the broader ecosystem, which now contributes $3 Trillion to Global GDP, provides the resources and finance to maintain quality and robust standards. Otherwise we could not make phone calls or mobile transactions.

ISOGG is well placed to play a key role in such an industry association for the Genetic Genealogy and Population Genetics sector. At some stage the industry will have to come together and replicate what many other industry associations have done to face similar challenges.

RCO
01-25-2016, 09:30 PM
We also have the minimal reference phylogeny for the human Y chromosome:
http://www.phylotree.org/Y/
They recognized my J1 FGC6064 SNP and I sincerely hope geneticists start to test people from the core areas of Northern Iran, Eastern Anatolia, Central Asia, Caspian Sea. ISOGG can include FGC6064. I like YFull tree as the best operational tree nowadays.

dp
01-25-2016, 10:54 PM
I used to use Phylotree for mtDNA branching. I just discovered it kept a Y-DNA phylogenetic tree last week :-)
dp

Huntergatherer1066
01-25-2016, 10:57 PM
We also have the minimal reference phylogeny for the human Y chromosome:
http://www.phylotree.org/Y/
They recognized my J1 FGC6064 SNP and I sincerely hope geneticists start to test people from the core areas of Northern Iran, Eastern Anatolia, Central Asia, Caspian Sea. ISOGG can include FGC6064. I like YFull tree as the best operational tree nowadays.

This is one of my favorite trees for a general overview, very straightforward and no frills. I generally try to check all of the haplotrees to create sort of a composite picture, they all have their strengths and weaknesses and draw from not entirely overlapping sample sets.

VinceT
01-25-2016, 11:00 PM
ISOGG is well placed to play a key role in such an industry association for the Genetic Genealogy and Population Genetics sector. At some stage the industry will have to come together and replicate what many other industry associations have done to face similar challenges.
Not so. ISOGG is not a standards organization nor is it an industry association. It is the brand name for a social network that is centered around the interests of a certain woman from southern California.

thetick
01-25-2016, 11:38 PM
Not so. ISOGG is not a standards organization nor is it an industry association. It is the brand name for a social network that is centered around the interests of a certain woman from southern California.

What should happen is define a preset of rules for adding to the tree. Have a policy where as prominent individuals and corporations have a vote in changing the rules. Define a discussion platform (Anthrogenica or rsm2 's WF are options). Open the doc under the GNU Free Documentation License and the document naturally will get filled in by competent passionate individuals. Also someone will need to offer a web service for the official tree (JOlson or Alex) ?

Personally I would adopt Alex's P312 bigtree as a place to start with Alex's participation and see if prominent individuals (Cofgene, GTC, lgmayka, Mikewww, Alex, VinceT, rms2, any active FTDNA Project Admin etc..and corporations (YFull, FTDNA, FGS) are willing to contribute.

Again this is very easy for me document what needs to be done as I have been involved in similar type projects but they involved computer software development. The actual coordinating , aligning and persaudinging is the very difficult part which I don't have to time or resources to devote outside of being an interested user.

GoldenHind
01-25-2016, 11:47 PM
We also have the minimal reference phylogeny for the human Y chromosome:
http://www.phylotree.org/Y/


Its not only minimal, it's also out of date. I note it doesn't include DF99, whose position as a subclade under P312 was established almost two and a half years ago.

thetick
01-25-2016, 11:51 PM
This is one of my favorite trees for a general overview, very straightforward and no frills. I generally try to check all of the haplotrees to create sort of a composite picture, they all have their strengths and weaknesses and draw from not entirely overlapping sample sets.

To be honest with you it's almost useless for those with NGS results. It does not even have SRY-2627, the most documented and studied group under R1b! Countless almost hundreds of groups missing under DF-27.

Huntergatherer1066
01-26-2016, 12:26 AM
To be honest with you it's almost useless for those with NGS results. It does not even have SRY-2627, the most documented and studied group under R1b! Countless almost hundreds of groups missing under DF-27.

Right, it's more of a trunk, without the branches and twigs.

RCO
01-26-2016, 01:03 AM
Yes and I was happy because at least the minimal included my FGC6064 SNP formed 15000 years ago ! We would need a new YCC 2016-2020 with international, institutional rules and control. The ancient people here can remember DNA-forums where we wrote important texts and suddenly it was gone because it was privately held in spite of several contributions. Imagine if a private company like FTDNA goes bankrupt in the future and the info could be threatened just like in any private enterprise, remember SMGF, for instance.

Cofgene
01-26-2016, 12:16 PM
Not so. ISOGG is not a standards organization nor is it an industry association. It is the brand name for a social network that is centered around the interests of a certain woman from southern California.


AMEN!

Peter M
01-27-2016, 09:42 AM
ISOGG's simple HTML based tree cannot possibly be expected to cope with the tsunami of SNPs flowing from NGS techniques.

I gather that there was some talk a year or more ago within ISOGG about converting it to a model with a database at its center, but nothing came of that.
But the idea was worked on. Katherine (Mrs. Borges) did as she always does: found a friend who was willing to help, for free of course, and kept her actions secret. The result was TAP (The Alleged Programmer) working on TAD (The Alleged Database). This framework (TAP working on TAD) was apparently instantiated multiple times (people in ISOGG are not co-operating) and I got in contact with one of those "leaders" who had done so (instantiate the TAP/TAD-framework), who sent me a first screenshot of the new ISOGG SNP Database. What I saw was mildly amusing to me and I told my contact his TAP didn't have the faintest idea (a) what he was doing; (b) what was/is really needed; and (c) the difference between the two. I warned him his TAP was not going anywhere and he had better stop his activities.

The fundamental flaw in his approach was, that he had referred his TAP to the ISOGG web site and asked him to put that (the tree on it) in a database, as that was all he had to show (the best he could do in the situation). And this web site does indeed show one of the biggest issues in tree management by the ISOGG i.e. keeping the SNP Index in synch with the tree. It's hard to believe, but this is actually being done by hand at the moment afaia, so every minor change to the tree is an issue and larger-scale changes require weeks. This flaw was of course easy to spot and so I told my contact about it and I explained to him him the easiest/best way out. But apparently he didn't believe me and some time later started complaining his TAP had stopped communication to him. 100% predictable; it was the only way out for the TAP without losing his face. All this (only) serves to show that (a) individuals in ISOGG are well aware they need a "system" to keep the Y-Tree Database, that (b) their html editor (together with the CSS3 sheet a friend made for them and Mrs. Borges claims to be co-author of) is not such a system and (c) are working on it (in their usual but completely ineffective way).

The biggest problem of ISOGG is that they seem to think (nobody is willing to discuss things with me, so I cannot be sure), building such a system is the work of "a programmer" and would take a few weeks or one or two months at most. This point is illustrated by "emails to all" sent out by ISOGG/Mrs. Borges recently and suggesting a new system/tree (distinction was unclear) to be available by Jan 1st 2016. Having done the activity (read on) I would like to meet this ISOGG hero. He must come directly from one of the Harry Potter movies and build systems using pure magic. My experience is, building such a system takes a number (>3-5) of man-years and requires a significantly extensive and diverse set of skills (the average programmer is highly unlikely to have or he would be making a fortune offering his services on the commercial market).

Personally, I think the activity should start with a complete reconsideration of the way the "Y-Tree" is organised, as currently the ISOGG tree is not resilient and fundamentally resists change. I tend to ignore the ISOGG tree, but recently saw the R-Z18 branch consists of about 6 SNPs. In reality (I administer the R-Z18 branch) there are well over 150 SNPs in that branch, in other words, 25 times more. I am convinced the tree ISOGG is showing on their web site will not survive e.g. a 25x increase in size of some branches without getting noticably inconsistent (it already is, but one needs my analysis software to detect it). There are (lots of) other factors, so the Maintenance if the Y-Tree should be reconsidered fundamentally from ground zero, e.g. uncoupling (a) the tree as such, (b) the display of the tree; and (c) the use of the tree as reference data in analysis tools (but I'm afraid this point is *FAR* too sophisticated for the current ISOGG; they will not be able to guide such a re-organisation.


10 years down the track I think the task is now well beyond the well-meaning ISOGG volunteers and I would prefer any such tree to be managed by professionals, as long as they were independent of the testing companies.Professionals required would be of a new type, I guess, as few people with have the designation "Branch Administrator" in their job description; and even if one person does, he would not have the background to define a Y-Tree database as the heart of the (business/functional/workflow/technical) architecture to support the complete field of Genetic Genealogy.

To get a discussion going, I posted a short description of my personal position in all of this on this forum (see Y-Tree-Database (http://www.anthrogenica.com/showthread.php?5914-Y-Tree-Database)). Unfortunately, the total number of serious responses did not significantly differ from zero and I left the discussion. Apparently nobody here is willing to discuss the definition of the Y-Tree Maintenance process, or not willing to discuss it in public, or not willing to discuss it with me (which implies with the R-Z18 Project) or whatever. Anyway, it's not happening. Read my post, I have a working system for Y-Tree maintenance running in the R-Z18 Project and we use it for all administration and analysis functions and software (the system is coupled to our analysis tools).

All of this convinced me that the biggest obstacle in defining an alternative to ISOGG, their stone-age Y-Tree Maintenance and their Development of Political Games (they only never officially make it to shops, which could be considered a pity; they would make great video games), is the apparent unwillingness of people to co-operate on this (not only with me but also with each other). This is by far the biggest obstacle, not the availability of talent or vision or whatever, but the unwillingness of people to co-operate, e.g. by setting up a new structure that convinces and/or allows all to make a contribution.

GoldenHind
01-27-2016, 07:36 PM
Good to have you back, Peter!