Andreas, I considered providing a Y database myself once, but I lack computer skills, so I finally came to my senses. For a time I even had a couple of web addresses on the back burner, I think I may still be paying for the domain names for some. I concluded it seemed like a significant undertaking. I presume it would have ongoing costs, but not sure what. Of the existing databases...... A) FTDNA CUSTOMER DATABASE: For the most part FTDNA are not interested in adding search features requested by customers, but in general they do moderately well so I am very thankful for what they provide. The search feature I would most like added would be ability to search for matches having my haplogroup branch SNPs, such as Z2552, L617, and FGC14951. The last one, FGC14951, is about 3,000 years old, and appears to denote a Bronze Age immigrant to England about 3,000 years ago. If I could trace more lines in England, it would help greatly in my efforts along with others to learn about this Bronze Age ancestor, and how the family spread around England before the "genealogical time frame" was reached about 800 years ago. The search feature of next interest to me would be widening the search criteria from for example a maximum of mutation 10 steps on 111 markers. At that level it prevents finding matches within my family group related in the past 300 to 800 years. It does not find any false matches, but false matches can be eliminated with research, so a few false matches could be tolerated if the matching included all of my close relatives. An issue though is that different haplogroups have very different numbers of matches, so one size may not fit all. The search feature of next greatest interest would be a way to overcome problems caused by multi step mutations on a single marker. In my family there was a 4 step mutation on one marker, which prevents a lot of matches being found. Some databases allow switching off multi copy markers in matching, as they can often make match finding difficult if that have recombination mutations The most important match for me on the FTDNA customer database is not contactable, but I guess that is their choice, but may just be an issue stemming from the person having his email address listed as his secondary email address rather than primary. B) YSEARCH: Biggest issue for me is that it does not include all of the markers 68 to 111, and it is in that group where 4 of my most important diagnostic family markers are, making it impossible to search YSearch for matches on my key family specific markers. The good thing about YSearch is that results can be uploaded direct from FTDNA, eliminating transcription errors. Without FTDNA cooperation, direct uploading to a new private database has a bit of a handicap. But YSeq might be amenable to direct upload features, as they would have something to gain from making finding matches easier for their customers. Perhaps FTDNA would cooperate.... if they are not worried about protecting their monopoly. Could ask them nicely. I think we are moving into an era where STRs need to be used in conjunction with SNPs, and known family history, and known geographic origins. I think that joining a database should have fields for earliest male line ancestor name, the country of origin of that ancestor, the county or state or province he came from, and the village or town, each individually searchable in the database. Perhaps scope for occupation, or other key words which might identify a family. It would be good if I could search for all persons tracing to my small ancestral village, as most in the village will be my relations on maternal lines within half a dozen generations. This would also help find matches from the area which had a surname change in recent centuries. There may be advantage in a person being able to insert a short biography of their earliest ancestor, say less than 50 or 100 words? I mentioned SNPs, and I think it would be good if a person could insert into a field at least their major haplogroup, plus their 3 or 4 most recent named SNPs as a minimum. In my case if I could search for matches using my recent SNPs as well as STRs I could focus in on important matches, and not be distracted with false matches from remote haplogroup branches. It might even be good if people tested on Y SNPs at Britains DNA, 23 and ME, national geographic etc could input their recent branch Y SNPs, even if they have not tested end STRs. Searches for surnames, surname variant, and combinations of surnames at the same time might be good. A new database may need an anonymous contacting system like YSearch has. When searching for matches, YSearch allows searching on a minimum of 8 random markers. That is good, but on the odd time it would be convenient if I could search with less markers. I think a new independent database should accommodate a minimum of the standard 111 FTDNA markers. But I now have found some diagnostic markers for my family in the many hundreds of new STR markers reported by Full Genomes Corp, and YFull, and YSeq. I have not yet determined the number of potentially useful STR markers, but just plucking a number out of the air, it might be 1000 or more. The new comprehensive genome tests often rate SNPs and STRs with reliability ratings, so you may not just have 1000 markers, but if you get fussy, you could at the very least have 2 reliability categories for each marker. The Captcha (spelling?) hindrance could be eliminated and save thousands of hours. I think the database should serve the community, and not be to fussy about protecting monopolies. Joining the database should have well though out privacy criteria which are well understood by all parties. There many more search features which could be added, which can be discussed if you get serious, and are not put off by my above comments. C) SEMARGL.ME: This is a very useful database largely captured from FTDNA surname projects and YSearch. Be sure to have a look at how that works. To quickly build a viable database.... How good would you be at following semargls lead??? There may be ethical issues here, or legal issues, not sure, so I am not trying to insight anything illegal or unethical, just noting that incorporating what information is currently public because FTDNA customers in projects have elected to make it public, would be something to think about. I don't believe persons can enter their own data or contact links onto SEMARGL, so that is a bit of a shortcoming. The SEMARGL database cannot be search on individual markers, or selections of markers, which is one of it's down sides. It can be searched for some major branch SNPs, but not many down stream SNPs. SEMARGL does limit searches to various max number of matches for different types of searches... I don't find this a major issue, just something to consider. Just a few of my thoughts! A last thought...... Perhaps the most important.... is "perminence". What can be done to make the database have a long life, 50 years or 100 years perhaps? Perhaps it needs to be run by a self renewing group rather than an individual, or at least have a built in succession feature to protect it in the future? John. Sent from my iPad > On 7/12/2015, at 9:26 pm, Andreas West via <genealogy-dna@rootsweb.com> wrote: > > Greg, > > I agree with your point that an alternative, company independent and open to all sorts of Y-DNA and mtDNA testing results would be good, if not needed (see the other now closed/destroyed databases at Ancestry). > > What kind of features would you expect from such a service? > > Comparing STR results sounds like an easy to feature, I have never looked at my results (not even sure if FTDNA allows for download of the raw Y-DNA data) but the resulting number of STR's must be very small in comparison to SNP's (given the larger effort to get them by hand). > > What else would you see as absolutely necessary, nice-to-have? That list should include features that might not exist today? > > I'm thinking about offering this maybe in the future but I haven't spend much time with it (Y-DNA) so I don't know how big of an effort it would be. > > Andreas > >> On 7 Dec 2015, at 03:47, G. Magoon via <genealogy-dna@rootsweb.com> wrote: >> >> The case based on database size is fine, from a practical standpoint; but >> in the absence of any other benefit, the argument basically boils down to: "I >> am recommending you test at FTDNA because I and most others have been >> recommending FTDNA previously." >> >> Personally, I think it would be stupendous if someone was able to set up an >> independent matching system (analogous to GEDMATCH) for Y-STRs (and mtDNA >> for that matter). With such a tool, prospective testers wouldn't have to >> worry so much about having to pay a significantly higher price in order to >> get that "ticket" into a large database. >> >> Your post also raises the question: how much is that "ticket" worth to a >> prospective tester? 50% extra cost? 100% higher cost? 200% higher cost? >> Personally, I think that is a decision that should be made by the >> prospective tester (or the person paying for the test), based on financial >> considerations, goals, etc. From my perspective, it is not as simple as >> "there is only one choice". >> >> I'm glad you raised the issue about results living on for posterity, which >> I also think is an important consideration. But I think testers at all >> labs, including FTDNA, would be wise to take appropriate safegaurds in this >> respect. An organization or business will typically only have a finite >> lifespan, and there are no guarantees that results will be >> hosted/maintained in perpetuity, FTDNA included. >> >> Again, just my two cents. >> >> Greg >> >> On Sun, Dec 6, 2015 at 1:37 PM, Mike W via <genealogy-dna@rootsweb.com> >> wrote: >> >>> Recruiting others to test, project and results sharing systems is critical. >>> This is the essence of genetic genealogy as comparisons between people and >>> sharing of results are needed. >>> >>> If possible, it is good in Y DNA testing to recruit get your most distant >>> male cousin to test so you can estimate the DNA information for the >>> family's most distant known ancestor (MDKA). However, it is hard to recruit >>> people on the other side of your genealogical brickwalls and you may not >>> want to pay for someone's tests you may not be related too. Ideally, you >>> want people who are inclined towards DNA testing to find you or for you to >>> find them through a matching database. >>> >>> The importance of a consistent set of test results available in a large >>> matching database can not be understated. This is FTDNA's most dominant >>> advantage for Y DNA, the largest accessible database. There hundreds of >>> thousands of Y DNA records already available in FTDNA's database. These are >>> records of real people, not just anonymous results. For more details, >>> please read this web page. >>> https://www.familytreedna.com/why-ftdna.aspx >>> >>> Cost is a factor, but when you get your Y STRs tested with FTDNA you are >>> also getting a ticket into that large database and project management >>> system with all of those surname projects. FTDNA allows you to join >>> projects without cost and there is no annual subscription fee for support >>> in their database and matching systems as well. >>> >>> Genetic test results data without the accompanying web based project and >>> matching systems is not as useful and may not live on for prosperity. When >>> you order STR and SNP tests from FTDNA your data is supported by a growing >>> company with a 15 year, self-sustaining operation. You don't have to send >>> your DNA sample to multiple labs. You and your recruits' samples all go to >>> Houston's lab and are stored there subject to published privacy policies. >>> This is particularly important if you think you only have one shot to get a >>> DNA sample from a recruit. >>> >>> The central lab and DNA storage support is complemented by a full product >>> line. Besides Y DNA testing up to 111 STRs, there is Y SNP testing and even >>> Y Next Generation SNP discovery testing (Big Y) along with an array of >>> autosomal and mitochondrial DNA tests. You have "one stop shopping" for >>> your DNA sample. >>> >>> I'm not known for political correctness and we do not want to scare you off >>> newbies but it is important that we acknowledge - It is very, very likely >>> that 37 Y STRs is not enough. I recommend you starting with a minimum of 67 >>> STRs. Most of the male large haplogroup branches of Europe started their >>> great expansions during the Bronze Age. That means that it is very hard to >>> discern who fits where at 37 STRs. 67 may not even be enough. Probably many >>> of the people posting here, even those who prefer niche vendors, have 67 >>> and even 111 STRs tested with FTDNA. >>> >>> Regards, >>> Mike W >>> >>> ------------------------------- >>> To unsubscribe from the list, please send an email to >>> GENEALOGY-DNA-request@rootsweb.com with the word 'unsubscribe' without >>> the quotes in the subject and the body of the message >> >> ------------------------------- >> To unsubscribe from the list, please send an email to GENEALOGY-DNA-request@rootsweb.com with the word 'unsubscribe' without the quotes in the subject and the body of the message > > > ------------------------------- > To unsubscribe from the list, please send an email to GENEALOGY-DNA-request@rootsweb.com with the word 'unsubscribe' without the quotes in the subject and the body of the message

Wow, that's a lot of features and I need to digest it further. I'm less worried about the different search features, that can be easily implemented, but the presentation of results and how to help filter out unimportant ones (nobody wants to go through 1500 matches by hand). If it's only 1000 STR that's easy, the number of SNP's is constantly growing and new SNP's are discovered with their importance and put into the phylotree (which is constantly changing). Thanks for giving such valuable input. A database cannot survive 50-100 years, technology is evolving to quickly, every 5 years there is a big step forward (as the processing power doubles every year our ability to analyze more and more data is growing as fast). Andreas > On 7 Dec 2015, at 17:20, AJ Marsh <ajmarshnz@gmail.com> wrote: > > Andreas, > > I considered providing a Y database myself once, but I lack computer skills, so I finally came to my senses. For a time I even had a couple of web addresses on the back burner, I think I may still be paying for the domain names for some. > > I concluded it seemed like a significant undertaking. I presume it would have ongoing costs, but not sure what. > > Of the existing databases...... > > A) FTDNA CUSTOMER DATABASE: > > For the most part FTDNA are not interested in adding search features requested by customers, but in general they do moderately well so I am very thankful for what they provide. > > The search feature I would most like added would be ability to search for matches having my haplogroup branch SNPs, such as Z2552, L617, and FGC14951. The last one, FGC14951, is about 3,000 years old, and appears to denote a Bronze Age immigrant to England about 3,000 years ago. If I could trace more lines in England, it would help greatly in my efforts along with others to learn about this Bronze Age ancestor, and how the family spread around England before the "genealogical time frame" was reached about 800 years ago. > > The search feature of next interest to me would be widening the search criteria from for example a maximum of mutation 10 steps on 111 markers. At that level it prevents finding matches within my family group related in the past 300 to 800 years. It does not find any false matches, but false matches can be eliminated with research, so a few false matches could be tolerated if the matching included all of my close relatives. An issue though is that different haplogroups have very different numbers of matches, so one size may not fit all. > > The search feature of next greatest interest would be a way to overcome problems caused by multi step mutations on a single marker. In my family there was a 4 step mutation on one marker, which prevents a lot of matches being found. Some databases allow switching off multi copy markers in matching, as they can often make match finding difficult if that have recombination mutations > > The most important match for me on the FTDNA customer database is not contactable, but I guess that is their choice, but may just be an issue stemming from the person having his email address listed as his secondary email address rather than primary. > > B) YSEARCH: > > Biggest issue for me is that it does not include all of the markers 68 to 111, and it is in that group where 4 of my most important diagnostic family markers are, making it impossible to search YSearch for matches on my key family specific markers. > > The good thing about YSearch is that results can be uploaded direct from FTDNA, eliminating transcription errors. Without FTDNA cooperation, direct uploading to a new private database has a bit of a handicap. But YSeq might be amenable to direct upload features, as they would have something to gain from making finding matches easier for their customers. Perhaps FTDNA would cooperate.... if they are not worried about protecting their monopoly. Could ask them nicely. > > I think we are moving into an era where STRs need to be used in conjunction with SNPs, and known family history, and known geographic origins. I think that joining a database should have fields for earliest male line ancestor name, the country of origin of that ancestor, the county or state or province he came from, and the village or town, each individually searchable in the database. Perhaps scope for occupation, or other key words which might identify a family. It would be good if I could search for all persons tracing to my small ancestral village, as most in the village will be my relations on maternal lines within half a dozen generations. This would also help find matches from the area which had a surname change in recent centuries. > > There may be advantage in a person being able to insert a short biography of their earliest ancestor, say less than 50 or 100 words? > > I mentioned SNPs, and I think it would be good if a person could insert into a field at least their major haplogroup, plus their 3 or 4 most recent named SNPs as a minimum. In my case if I could search for matches using my recent SNPs as well as STRs I could focus in on important matches, and not be distracted with false matches from remote haplogroup branches. It might even be good if people tested on Y SNPs at Britains DNA, 23 and ME, national geographic etc could input their recent branch Y SNPs, even if they have not tested end STRs. > > Searches for surnames, surname variant, and combinations of surnames at the same time might be good. > > A new database may need an anonymous contacting system like YSearch has. > > When searching for matches, YSearch allows searching on a minimum of 8 random markers. That is good, but on the odd time it would be convenient if I could search with less markers. > > I think a new independent database should accommodate a minimum of the standard 111 FTDNA markers. But I now have found some diagnostic markers for my family in the many hundreds of new STR markers reported by Full Genomes Corp, and YFull, and YSeq. I have not yet determined the number of potentially useful STR markers, but just plucking a number out of the air, it might be 1000 or more. The new comprehensive genome tests often rate SNPs and STRs with reliability ratings, so you may not just have 1000 markers, but if you get fussy, you could at the very least have 2 reliability categories for each marker. > > The Captcha (spelling?) hindrance could be eliminated and save thousands of hours. I think the database should serve the community, and not be to fussy about protecting monopolies. > > Joining the database should have well though out privacy criteria which are well understood by all parties. > > There many more search features which could be added, which can be discussed if you get serious, and are not put off by my above comments. > > C) SEMARGL.ME: > > This is a very useful database largely captured from FTDNA surname projects and YSearch. Be sure to have a look at how that works. To quickly build a viable database.... How good would you be at following semargls lead??? There may be ethical issues here, or legal issues, not sure, so I am not trying to insight anything illegal or unethical, just noting that incorporating what information is currently public because FTDNA customers in projects have elected to make it public, would be something to think about. I don't believe persons can enter their own data or contact links onto SEMARGL, so that is a bit of a shortcoming. > > The SEMARGL database cannot be search on individual markers, or selections of markers, which is one of it's down sides. It can be searched for some major branch SNPs, but not many down stream SNPs. SEMARGL does limit searches to various max number of matches for different types of searches... I don't find this a major issue, just something to consider. > > Just a few of my thoughts! A last thought...... Perhaps the most important.... is "perminence". What can be done to make the database have a long life, 50 years or 100 years perhaps? Perhaps it needs to be run by a self renewing group rather than an individual, or at least have a built in succession feature to protect it in the future? > > John. > > > > Sent from my iPad > >> On 7/12/2015, at 9:26 pm, Andreas West via <genealogy-dna@rootsweb.com> wrote: >> >> Greg, >> >> I agree with your point that an alternative, company independent and open to all sorts of Y-DNA and mtDNA testing results would be good, if not needed (see the other now closed/destroyed databases at Ancestry). >> >> What kind of features would you expect from such a service? >> >> Comparing STR results sounds like an easy to feature, I have never looked at my results (not even sure if FTDNA allows for download of the raw Y-DNA data) but the resulting number of STR's must be very small in comparison to SNP's (given the larger effort to get them by hand). >> >> What else would you see as absolutely necessary, nice-to-have? That list should include features that might not exist today? >> >> I'm thinking about offering this maybe in the future but I haven't spend much time with it (Y-DNA) so I don't know how big of an effort it would be. >> >> Andreas >> >>> On 7 Dec 2015, at 03:47, G. Magoon via <genealogy-dna@rootsweb.com> wrote: >>> >>> The case based on database size is fine, from a practical standpoint; but >>> in the absence of any other benefit, the argument basically boils down to: "I >>> am recommending you test at FTDNA because I and most others have been >>> recommending FTDNA previously." >>> >>> Personally, I think it would be stupendous if someone was able to set up an >>> independent matching system (analogous to GEDMATCH) for Y-STRs (and mtDNA >>> for that matter). With such a tool, prospective testers wouldn't have to >>> worry so much about having to pay a significantly higher price in order to >>> get that "ticket" into a large database. >>> >>> Your post also raises the question: how much is that "ticket" worth to a >>> prospective tester? 50% extra cost? 100% higher cost? 200% higher cost? >>> Personally, I think that is a decision that should be made by the >>> prospective tester (or the person paying for the test), based on financial >>> considerations, goals, etc. From my perspective, it is not as simple as >>> "there is only one choice". >>> >>> I'm glad you raised the issue about results living on for posterity, which >>> I also think is an important consideration. But I think testers at all >>> labs, including FTDNA, would be wise to take appropriate safegaurds in this >>> respect. An organization or business will typically only have a finite >>> lifespan, and there are no guarantees that results will be >>> hosted/maintained in perpetuity, FTDNA included. >>> >>> Again, just my two cents. >>> >>> Greg >>> >>> On Sun, Dec 6, 2015 at 1:37 PM, Mike W via <genealogy-dna@rootsweb.com> >>> wrote: >>> >>>> Recruiting others to test, project and results sharing systems is critical. >>>> This is the essence of genetic genealogy as comparisons between people and >>>> sharing of results are needed. >>>> >>>> If possible, it is good in Y DNA testing to recruit get your most distant >>>> male cousin to test so you can estimate the DNA information for the >>>> family's most distant known ancestor (MDKA). However, it is hard to recruit >>>> people on the other side of your genealogical brickwalls and you may not >>>> want to pay for someone's tests you may not be related too. Ideally, you >>>> want people who are inclined towards DNA testing to find you or for you to >>>> find them through a matching database. >>>> >>>> The importance of a consistent set of test results available in a large >>>> matching database can not be understated. This is FTDNA's most dominant >>>> advantage for Y DNA, the largest accessible database. There hundreds of >>>> thousands of Y DNA records already available in FTDNA's database. These are >>>> records of real people, not just anonymous results. For more details, >>>> please read this web page. >>>> https://www.familytreedna.com/why-ftdna.aspx >>>> >>>> Cost is a factor, but when you get your Y STRs tested with FTDNA you are >>>> also getting a ticket into that large database and project management >>>> system with all of those surname projects. FTDNA allows you to join >>>> projects without cost and there is no annual subscription fee for support >>>> in their database and matching systems as well. >>>> >>>> Genetic test results data without the accompanying web based project and >>>> matching systems is not as useful and may not live on for prosperity. When >>>> you order STR and SNP tests from FTDNA your data is supported by a growing >>>> company with a 15 year, self-sustaining operation. You don't have to send >>>> your DNA sample to multiple labs. You and your recruits' samples all go to >>>> Houston's lab and are stored there subject to published privacy policies. >>>> This is particularly important if you think you only have one shot to get a >>>> DNA sample from a recruit. >>>> >>>> The central lab and DNA storage support is complemented by a full product >>>> line. Besides Y DNA testing up to 111 STRs, there is Y SNP testing and even >>>> Y Next Generation SNP discovery testing (Big Y) along with an array of >>>> autosomal and mitochondrial DNA tests. You have "one stop shopping" for >>>> your DNA sample. >>>> >>>> I'm not known for political correctness and we do not want to scare you off >>>> newbies but it is important that we acknowledge - It is very, very likely >>>> that 37 Y STRs is not enough. I recommend you starting with a minimum of 67 >>>> STRs. Most of the male large haplogroup branches of Europe started their >>>> great expansions during the Bronze Age. That means that it is very hard to >>>> discern who fits where at 37 STRs. 67 may not even be enough. Probably many >>>> of the people posting here, even those who prefer niche vendors, have 67 >>>> and even 111 STRs tested with FTDNA. >>>> >>>> Regards, >>>> Mike W >>>> >>>> ------------------------------- >>>> To unsubscribe from the list, please send an email to >>>> GENEALOGY-DNA-request@rootsweb.com with the word 'unsubscribe' without >>>> the quotes in the subject and the body of the message >>> >>> ------------------------------- >>> To unsubscribe from the list, please send an email to GENEALOGY-DNA-request@rootsweb.com with the word 'unsubscribe' without the quotes in the subject and the body of the message >> >> >> ------------------------------- >> To unsubscribe from the list, please send an email to GENEALOGY-DNA-request@rootsweb.com with the word 'unsubscribe' without the quotes in the subject and the body of the message