> Where am I going wrong? > > Bruce Bruce, One thing that may be (probably?) throwing you is thinking of the Modal as THE ancestral haplotype. It could be...it is probably very close in many/most cases when you have a large sampling. But, you're never sure there's not an unintended bias such as the lines carrying the ancestral value for a particular marker not being as prolific. From your description, I got the following values: Individual DYS461 DYS449 MODAL 13 29 Indv #1 12 29 Indv #2 13 28 Indv #3 12 28 If, in the above example, the Ancestral haplotype was in actuality DYS461/449 = 13/28 or 12/29 your need for a parallel mutation goes away. If 13/28, individual #2's line would still be carrying the ancestral haplotype. A single mutation at DYS461 would account for individual #3 and a single mutation at DYS449 would account for individual #1. Both would be one step from the ancestral line. With the 12/29 scenario, individual #1 would be carrying the ancestral haplotype. A mutation at DYS449 takes you to individual #3 and then a mutation at DYS461 could take you to individual #2. Individual 3 would be 1 step from the ancestral line while #2 would be 2 steps away. I tend to like the simplest answers. While parallel mutations can and I'm sure do take place, absent a paper trail, I would be inclined to go with one of the two modal values not being ancestral. David M

Thanks David. I must be messing up on the definition of parallel mutations. All I can find is discussions in the GENEALOGY-DNA archives, and nothing mentions that one or the other mutation had to occur first, just that the mutations are independent. Now, it also mentions that the mutations must occur in individuals in two branches of a well documented family. I don't have that. All I have is that the individuals have the surname Baird, are the same Haplogroup and match exactly on a lot of other markers. I have a modal and three individuals, 1, 2, and 3. A mutation in DYS461 from 13 to 12 takes me from the modal to 1. A mutation in DYS449 from 29 to 28 takes me from the modal to 2. Now, another mutation of DYS449 from 29 to 28 will take me from 1 to 3 or another mutation from DYS461 will take me from 2 to 3. In either way I can only "connect all the dots" with two mutations of DYS461 or two mutation of DYS449. If the family was well documented and well tested, I would know which set of two mutations would exist and the Fluxus pattern would not close. For example, if I knew that 3 was descended from 1, the pattern would show the modal connected to 1 because of a mutation in DYS461 from 13 to 12, 1 connected to 3 because of a mutation in DYS449 from 29 to 28, and the modal connected to 2, also because of a mutation in DYS449 from 29 to 28. Thus, the parallel mutation. Where am I going wrong? Bruce In a message dated 3/27/2006 3:09:22 PM Central Standard Time, dmcduke@comcast.net writes: Bruce, Your use of the word 'alternate' is much better than 'parallel' mutations. What you've got is the Fluxus program not able to determine the path which was actually taken. If you have two mutations between point A and point B, there's no way to determine which took place first. David M

SCOT-DNA-L@rootsweb.com: In studying our growing R320 DNA project results, notice two "Haplotype" categories seem dominant for those of known Scot or Scot-Irish ancestry: "I" and "R1b1" There is also "R1b" and "R1b1c" represented and one new report indicating "K2." A couple of the project members have ordered the "SNP" analysis which confirmed "I" and R1b1." Questions: Are these the more common Haplos for Scotland? What does the SNP analysis report really tell us about our origins? What is the significance of matches to other surnames (most numerous seem to be several 37 markers exact matches to Bingham/Bigham/Bigum)? The genealogy DNA firms represented are FTDNA and EA (EthnoAncestry). Our project page also includes those with Germanic origins: <http://www.familytreedna.com/public/Reddick-r320/index.aspx>http://www.familytreedna.com/public/Reddick-r320/index.aspx Thanks, Richard D. Reddick

Hi David, Thanks for the help. Your suggestions worked. The verbiage in the box is a green check mark followed by "Finalize." I tried to open the *.bmp file with Adobe Photoshop 5.0 LE, Picture Publisher 6.0, and Imaging without success, but Paint worked. I saved it as a gif file which I opened with my browser. I also have several participants who tested with Sorenson. I encourage testing with them on my surname's mailing list since it's free. As a result my charts are either the 39 markers from SMGF or the 28 common markers between SMGF and the FTDNA 37 marker test. I started off by inputting all the marker data, but then realized that Fluxus only worked on mutations. So now I just enter data for those markers which have mutations. I also just enter data for a single haplogroup cluster for each plot and I also enter the modal values for the mutated markers as one record. This gives me a "feel" for determining how the mutations propagate, from the mode to the individual participants. One interesting thing about the Fluxus program are the alternate paths to a given participant which contain two identical mutations. These appear to me to be parallel mutations. I posted this on the Genealogy-DNA mailing list and Ann Turner said she would take a look at my plots, so I sent her the *.fdi plots. Evidently she couldn't read them, because she hasn't responded. I'll try again with the *.gif files. Bruce Baird In a message dated 3/26/2006 2:29:35 PM Central Standard Time, dmcduke@comcast.net writes: Bruce, See my notes below... > Hi David, > > The drawings I created using the latest Fluxus software show the marker > mutations on the lines and unnamed nodes at the points where the lines > connect. > Did you remove these? Yes, that is an option in the program. For the diagram to be readable on a website, you do need to get rid of that clutter. When you initially draw the network, if you scroll all the way down, you should see a box towards the right you can click. Unfortunately, I can't tell you what verbage may be in the box because on my screen I can't see it...haven't figured out why, yet. I can see the upper edge and when I click on it, display parameters pop up. I uncheck 'Display Mutated Positions' and 'Show Median Vectors' before saving the image as a .bmp file type. I also adjust the font. >Also, is there a program for converting the Fluxus > formatted diagram (*.fdi) file to a gif file like the one on your web > site? Don't save the file as *.fdi if you want to eventually save it as a *.gif. Save it as a *.bmp (an option) and then you're able to manipulate the file in a program such as Paint, Adobe, etc. Whatever program you may have that will open a *.bmp and then allow you to edit, add text, etc. Once you've edited the file you can then save it as a *.gif for use on a website. I'm still new to the software myself and found out just today that I could add the color coding to the nodes by right clicking the node I want to change. Various options come up, one being change of color. I had been doing that using one of the image editors. > > Bruce Baird

Brian, Thank you for your comments. I have modified the utility to post a warning if there are any missing values in the set of enabled markers when Fluxus output is enabled. The warning recommends to either disable the offending markers or remove offending haplotypes. Also D3892 has been renamed to D389B. -Dean > From: Bbairdsr@aol.com [mailto:Bbairdsr@aol.com] > > Regarding the handling of missing markers, possibly it would > be better to > post a warning message when missing markers are encountered > instead of displaying > the ych file. Fluxus turned itself off when I tried to run a > file with a lot > of zeros in it. I had run a small sample file with no > problem so I couldn't > figure out what the problem was. I even posted a question on the > Genealogy-DNA mailing list about this, but had no responses. > Then I figured it must have > been all the zeros, and it was. I don't know of an > acceptable format for > missing markers that Fluxus will accept.

Hi Dean, I apologize for the DYS389i and DYS389ii comment. I checked the ych file that your program produced and it had converted the value of DYS389ii. I think what threw me off was that the name of D3892 remained the same instead of being changed to something like D389B, which SMGF uses. Regarding the handling of missing markers, possibly it would be better to post a warning message when missing markers are encountered instead of displaying the ych file. Fluxus turned itself off when I tried to run a file with a lot of zeros in it. I had run a small sample file with no problem so I couldn't figure out what the problem was. I even posted a question on the Genealogy-DNA mailing list about this, but had no responses. Then I figured it must have been all the zeros, and it was. I don't know of an acceptable format for missing markers that Fluxus will accept. I did use your feature for disabling markers and it worked great, like everything else in your program. Thank you so much for developing this software for us would be genetic genealogists. I wish I could produce such nice looking HTML tables from Excel 97 using Microsoft's conversion program. Bruce Baird In a message dated 3/26/2006 7:12:34 PM Central Standard Time, hdmcgee000@comcast.net writes: Bruce, Yes, missing markers are a problem for making cladograms with the Y-Utility. The instructions indicate that the "haplotype data for all sets of results have the same number of markers". If you have a suggestion for a better way of handling missing markers, please let me know. Is there a format accepted by the Fluxus software that accounts for missing markers? The Y-Utility has been upgraded with the capability of enabling each marker. So, you can disable the markers which exist for only part of the set of results. Then the .ych file will only contain the values for enabled markers. Regarding DYS389i and DYS389ii, these are handled properly. The Y-Utility for your example would convert DYS389ii to the value of 17 and leave the values of 12 and 13 for DYS389i, so there would only be one mutation. The colors in the displayed tables also handle this marker properly, although the displayed value of DYS389ii is not changed. I would be interested in any suggestions for a better way to handle DYS464. Please let me know if the Y-Utility is not working properly in any way. -Dean McGee

> From: David [mailto:dmcduke@comcast.net] > > By the way, Dean...are you planning on updating your utility > to account for > the additional markers that FTDNA has added? I hope so but do > understand if > you don't. Yes, I plan on keeping it compatible with copy-and-paste from Ysearch and from FTDNA generated web pages. I assume that FTDNA will update these tables when the results start rolling in.

> Please let me know if the Y-Utility is not working properly in any way. > > -Dean McGee > > Dean, Your utility works very well. Thank you!!! It's the limitations of the Fluxus software that we have to work around. Whether you were to put a space or the zero as you currently do, the Fluxus software interprets it as a zero. It would then interpret the zero as a mutation compared to a taxon (haplotype) that had a value for that marker which skews the results. Bottom line is that you can't mix no data with data and get good results. By the way, Dean...are you planning on updating your utility to account for the additional markers that FTDNA has added? I hope so but do understand if you don't. David M

Bruce, Yes, missing markers are a problem for making cladograms with the Y-Utility. The instructions indicate that the "haplotype data for all sets of results have the same number of markers". If you have a suggestion for a better way of handling missing markers, please let me know. Is there a format accepted by the Fluxus software that accounts for missing markers? The Y-Utility has been upgraded with the capability of enabling each marker. So, you can disable the markers which exist for only part of the set of results. Then the .ych file will only contain the values for enabled markers. Regarding DYS389i and DYS389ii, these are handled properly. The Y-Utility for your example would convert DYS389ii to the value of 17 and leave the values of 12 and 13 for DYS389i, so there would only be one mutation. The colors in the displayed tables also handle this marker properly, although the displayed value of DYS389ii is not changed. I would be interested in any suggestions for a better way to handle DYS464. Please let me know if the Y-Utility is not working properly in any way. -Dean McGee > -----Original Message----- > From: Bbairdsr@aol.com [mailto:Bbairdsr@aol.com] > > In a message dated 3/26/2006 3:00:41 PM Central Standard Time, > dmcduke@comcast.net writes: > A much easier way, is by using Dean McGee's Y-DNA Comparison > Utility... > http://www.mymcgee.com/tools/yutility.html . You'll still > need to have your > haplotype data in a spreadsheet format in order to load it > into the utility > but it does the conversion to the *.ych format for you. > Dean McGee's Y-DNA Comparison Utility substitutes a zero for > any marker value > that is blank, because of lack of data. I initially > converted a 23 person, > 48 marker spreadsheet with a lot of blank spaces using his > utility, then ran > Fluxus. The program turned itself off. I then tried a > simple case with one > zero for a marker that was 22 for other subjects and it drew > 22 mutations. So > make sure all your cells have data. Also, McGee's program > converts FTDNA > DYS389i and DYS389ii data as it stands and doesn't convert to > DYS389i and DYS389B. > So Fluxus thinks there are two mutations between a 12, 29 set > and a 13, 30 set > instead of one. I imagine it would have problems with the > DYS464 data also. > > Bruce Baird

> In a message dated 3/26/2006 3:00:41 PM Central Standard Time, > dmcduke@comcast.net writes: > A much easier way, is by using Dean McGee's Y-DNA Comparison Utility... > http://www.mymcgee.com/tools/yutility.html . You'll still need to have > your > haplotype data in a spreadsheet format in order to load it into the > utility > but it does the conversion to the *.ych format for you. > Dean McGee's Y-DNA Comparison Utility substitutes a zero for any marker > value > that is blank, because of lack of data. I initially converted a 23 > person, > 48 marker spreadsheet with a lot of blank spaces using his utility, then > ran > Fluxus. The program turned itself off. I then tried a simple case with > one > zero for a marker that was 22 for other subjects and it drew 22 mutations. > So > make sure all your cells have data. Also, McGee's program converts FTDNA > DYS389i and DYS389ii data as it stands and doesn't convert to DYS389i and > DYS389B. > So Fluxus thinks there are two mutations between a 12, 29 set and a 13, 30 > set > instead of one. I imagine it would have problems with the DYS464 data > also. > > Bruce Baird Both very good points. I'm usually deleting the DYS464 data before running it through Fluxus because we have several participants who tested with Sorenson and they don't test for that marker. I also make sure that I have a like number of tested markers for all the participants I run through the program. I've never bothered running the 12 marker haplotype through because I just don't see the use in it. With the FTDNA 25 marker results, I'm running 21 markers through Fluxus...25 minus the DYS464's. At the FTDNA 37 level, I'm taking out the DYS464's as well as DYS607, 576,570, and CDYa&b, which are not tested by Sorenson, so I'm actually running 29 markers through. You do need to make sure whatever markers you are using are all populated for all haplotypes to get good results. David M

In a message dated 3/26/2006 3:00:41 PM Central Standard Time, dmcduke@comcast.net writes: A much easier way, is by using Dean McGee's Y-DNA Comparison Utility... http://www.mymcgee.com/tools/yutility.html . You'll still need to have your haplotype data in a spreadsheet format in order to load it into the utility but it does the conversion to the *.ych format for you. Dean McGee's Y-DNA Comparison Utility substitutes a zero for any marker value that is blank, because of lack of data. I initially converted a 23 person, 48 marker spreadsheet with a lot of blank spaces using his utility, then ran Fluxus. The program turned itself off. I then tried a simple case with one zero for a marker that was 22 for other subjects and it drew 22 mutations. So make sure all your cells have data. Also, McGee's program converts FTDNA DYS389i and DYS389ii data as it stands and doesn't convert to DYS389i and DYS389B. So Fluxus thinks there are two mutations between a 12, 29 set and a 13, 30 set instead of one. I imagine it would have problems with the DYS464 data also. Bruce Baird

> R1a is more rarer than R1b > > keep on-keeping on-never quit. Roger I'm not sure if you're making a statement or asking a question. In western Europe, R1b is the most numerous haplogroup though someone out there may know of specific/localized examples of where that would not be the case and R1a might be more common in those specific situations. Without knowing context, it's impossible to answer. David M

>I don't understand how Fluxus works, and can't find anything on their > website that explains it. > > a. Did you input all the DYS values for all R1b Clan MacMillan > participants > and then the software supplied ancestral data and linkages back through > time? > b. If so, where did the software get that ancestral data? > > -Bill Martin Denver, CO Bill, There is no 'ancestral' data unless you want to call Y-DNA STRs ancestral. What Fluxus essentially does is represent, in a visual manner, the relative relatedness of the haplotypes you input. It's based on the mutations/differences between the haplotypes but how it actually does it, I couldn't begin to tell you. I have seen some links on their website to what I expect are academic papers explaining some of the workings/ideas behind the software. I've not taken the time to really delve into it. Sort of like I know basically what a TV does but I really have no idea how to build/fix one. What I see coming out of the Fluxus software corresponds very closely [I'd say exactly really] to what I've done using the calculations of TMRCA's and genetic distance when grouping the results for Clan MacMillan. I like the fact that my initial grouping are being validated by a second method. For question 'a' above: I maintain the Clan MacMillan results in a spreadsheet. Initially, I was converting the Excel file to a comma delimited file, doing some other clean up, adding required headers etc. and then saving the results as a *.ych file. That is an error prone way but I managed by being careful and the fact I deal with file manipulation in my job, daily. A much easier way, is by using Dean McGee's Y-DNA Comparison Utility... http://www.mymcgee.com/tools/yutility.html . You'll still need to have your haplotype data in a spreadsheet format in order to load it into the utility but it does the conversion to the *.ych format for you. For question 'b' above: all data is coming from my input. David M

R1a.any information on this group helpfull.? keep on-keeping on-never quit. Roger http://community.webtv.net/zgordo/GORDOSGENEALOGY

Bruce, See my notes below... > Hi David, > > The drawings I created using the latest Fluxus software show the marker > mutations on the lines and unnamed nodes at the points where the lines > connect. > Did you remove these? Yes, that is an option in the program. For the diagram to be readable on a website, you do need to get rid of that clutter. When you initially draw the network, if you scroll all the way down, you should see a box towards the right you can click. Unfortunately, I can't tell you what verbage may be in the box because on my screen I can't see it...haven't figured out why, yet. I can see the upper edge and when I click on it, display parameters pop up. I uncheck 'Display Mutated Positions' and 'Show Median Vectors' before saving the image as a .bmp file type. I also adjust the font. >Also, is there a program for converting the Fluxus > formatted diagram (*.fdi) file to a gif file like the one on your web > site? Don't save the file as *.fdi if you want to eventually save it as a *.gif. Save it as a *.bmp (an option) and then you're able to manipulate the file in a program such as Paint, Adobe, etc. Whatever program you may have that will open a *.bmp and then allow you to edit, add text, etc. Once you've edited the file you can then save it as a *.gif for use on a website. I'm still new to the software myself and found out just today that I could add the color coding to the nodes by right clicking the node I want to change. Various options come up, one being change of color. I had been doing that using one of the image editors. > > Bruce Baird Dave > > > ==== SCOT-DNA Mailing List ==== > All posts to this list are archived and cannot be edited from: > http://archiver.rootsweb.com/th/index/SCOT-DNA/ > Please bear this in mind if you are considering posting > anything of a sensitive nature re your personal DNA. > > ============================== > Search the US Census Collection. Over 140 million records added in the > last 12 months. Largest online collection in the world. Learn more: > http://www.ancestry.com/s13965/rd.ashx >

Many thanks, David. Now I've got a much clearer understanding of the process. -Bill ----- Original Message ----- From: "David" <dmcduke@comcast.net> To: <SCOT-DNA-L@rootsweb.com> Sent: Sunday, March 26, 2006 2:00 PM Subject: Re: [SCOT-DNA] cladogram > >I don't understand how Fluxus works, and can't find anything on their > > website that explains it. > > > > a. Did you input all the DYS values for all R1b Clan MacMillan > > participants > > and then the software supplied ancestral data and linkages back through > > time? > > b. If so, where did the software get that ancestral data? > > > > -Bill Martin Denver, CO > > Bill, > > There is no 'ancestral' data unless you want to call Y-DNA STRs ancestral. > What Fluxus essentially does is represent, in a visual manner, the relative > relatedness of the haplotypes you input. It's based on the > mutations/differences between the haplotypes but how it actually does it, I > couldn't begin to tell you. I have seen some links on their website to what > I expect are academic papers explaining some of the workings/ideas behind > the software. I've not taken the time to really delve into it. Sort of like > I know basically what a TV does but I really have no idea how to build/fix > one. > > What I see coming out of the Fluxus software corresponds very closely [I'd > say exactly really] to what I've done using the calculations of TMRCA's and > genetic distance when grouping the results for Clan MacMillan. I like the > fact that my initial grouping are being validated by a second method. > > For question 'a' above: I maintain the Clan MacMillan results in a > spreadsheet. Initially, I was converting the Excel file to a comma delimited > file, doing some other clean up, adding required headers etc. and then > saving the results as a *.ych file. That is an error prone way but I managed > by being careful and the fact I deal with file manipulation in my job, > daily. > > A much easier way, is by using Dean McGee's Y-DNA Comparison Utility... > http://www.mymcgee.com/tools/yutility.html . You'll still need to have your > haplotype data in a spreadsheet format in order to load it into the utility > but it does the conversion to the *.ych format for you. > > For question 'b' above: all data is coming from my input. > > David M > > > ==== SCOT-DNA Mailing List ==== > Have questions about lab cost? Contact the Project Manager, > John A. Hansen, directly at dnaclans@brigadoon.net and he will > provide a private answer. > Want to join the Project? Visit: http://www.ftdna.com/surname_det.asp?group=Scottish-Clans&projecttype=G > > ============================== > Search the US Census Collection. Over 140 million records added in the > last 12 months. Largest online collection in the world. Learn more: http://www.ancestry.com/s13965/rd.ashx >

R1a is more rarer than R1b keep on-keeping on-never quit. Roger http://community.webtv.net/zgordo/GORDOSGENEALOGY

In a message dated 3/26/2006 9:05:40 AM Central Standard Time, dmcduke@comcast.net writes: I've used that software to produce a diagram showing the Clan MacMillan R1b participants. If you're interested in what the output may look like, take a look at http://home.comcast.net/~mcmillanmail/DNA/Pages/Diagram.html . Click on the diagram to see a full sized version. David McMillan Hi David, The drawings I created using the latest Fluxus software show the marker mutations on the lines and unnamed nodes at the points where the lines connect. Did you remove these? Also, is there a program for converting the Fluxus formatted diagram (*.fdi) file to a gif file like the one on your web site? Bruce Baird

I don't understand how Fluxus works, and can't find anything on their website that explains it. a. Did you input all the DYS values for all R1b Clan MacMillan participants and then the software supplied ancestral data and linkages back through time? b. If so, where did the software get that ancestral data? -Bill Martin Denver, CO ----- Original Message ----- From: <Bbairdsr@aol.com> To: <SCOT-DNA-L@rootsweb.com> Sent: Sunday, March 26, 2006 10:41 AM Subject: Re: [SCOT-DNA] cladogram > In a message dated 3/26/2006 9:05:40 AM Central Standard Time, > dmcduke@comcast.net writes: > I've used that software to produce a diagram showing the Clan MacMillan R1b > participants. If you're interested in what the output may look like, take a > look at http://home.comcast.net/~mcmillanmail/DNA/Pages/Diagram.html . Click > on the diagram to see a full sized version. > > David McMillan > Hi David, > > The drawings I created using the latest Fluxus software show the marker > mutations on the lines and unnamed nodes at the points where the lines connect. > Did you remove these? Also, is there a program for converting the Fluxus > formatted diagram (*.fdi) file to a gif file like the one on your web site? > > Bruce Baird > > > ==== SCOT-DNA Mailing List ==== > All posts to this list are archived and cannot be edited from: > http://archiver.rootsweb.com/th/index/SCOT-DNA/ > Please bear this in mind if you are considering posting > anything of a sensitive nature re your personal DNA. > > ============================== > Search the US Census Collection. Over 140 million records added in the > last 12 months. Largest online collection in the world. Learn more: http://www.ancestry.com/s13965/rd.ashx >

> >I was just reading Family Tree magazine (April 2006). There's free > >software to make a cladogram (aka phylogenetic tree). I'm not exactly > >sure that I can explain it, but if you're interested, this is the link > >for Network software: >> http://www.fluxus-engineering.com/sharenet.htm >> >> ---Robin M > > Robin, > > I've used that software to produce a diagram showing the Clan MacMillan > R1b participants. If you're interested in what the output may look like, > take a look at > http://home.comcast.net/~mcmillanmail/DNA/Pages/Diagram.html . Click on > the diagram to see a full sized version. > > David McMillan Actually, I misspoke. It includes all the participants with the FTDNA 25 Y-DNA markers tested. R1b is towards the top and our I (indicated I) participants are at the bottom. Group 1 (indicated by red) includes Chief George MacMillan of MacMillan and Knapp. David M