Login to participate
  
Register   Lost ID/password?

Louis Kessler's Behold Blog

Double Match Theorem 1 - 23 hrs, 8 min ago

When I think of something in the middle of the night, I just have to write it down as soon as I get up. My mind’s background processor has been absorbed in figuring out how to get the output from Double Match Triangulator to display the two halves of each Chromosome pair.

It’s tricky. Currently, DMT is determining Double Match Groups by looking for overlapping matching segments. These Double Match Groups themselves may overlap and you can have groups within groups. Any relative used as Person b in Double Matching may match to Person a on both halves, meaning they are connected in more than one way, both of them through both their parents.

What I realized last night, and what I’m going to present as a Theorem (going back to my mathematical days) is:

Double Match Theorem 1: Any segment where Person a and Person b Triangulate with any other person must be on the other half of the Chromosome pair of any overlapping Missing a-b segments.

Well that’s a bit of gobbledygook. What it means is that I might be able to start separating out Person a’s Chromosome pairs. Take a look for example at the following map produced by DMT:

image

To refresh your memory, the green X’s are Double Matches, where both Person a matches Person c and Person b matches Person c. The pink a’s are where only Person a matches Person c, and the blue b’s are where only Person b matches Person c.

The green X’s in the yellow line are where Person a matches Person b. Any Double Matches that overlap will have the third leg needed to Triangulate. Those that do are designated as “Full Triangulation”.

Notice there are 7 lines that are marked as Missing a-b Match. Their Double Match area (two green X’s) end in the 142,000,000 to 142,999,999 address range just before the Base a-b begins. Their Double Match area does not intersect with the a-b match and therefore they do not Triangulate.

However, notice that 5 of the matches have those pink a’s extending to the right into the Triangulation area. This means that Person a matches Person c over that area. But Person b does not match Person c over that area or those would not have been pink a’s but green X’s. Yet, we know that Person a matches Person b on one of the Chromosome pair over that region. Therefore, these matches must be on the other Chromosome of the pair.  Q.E.D.

(Wow. I don’t think I’ve used Q.E.D. at the end of a theorem in like 40 years)

As a result of this, I am going to separate the Full Triangulations from the Missing a-b Matches prior to the determination of Double Match Groups. This will result in pure Triangulation Groups and pure Double Match Groups. I’ll have to see what this does and how it works and after a bit of experience with it, maybe it will lead to the possible enhancement and ultimately to full identification of both halves of each Chromosome pair.

And while I’m at it, let’s bring up another term from my early mathematical days:

Double Match Corollary 1: Any segment where Person a and Person b Triangulate on both halves of the Chromosome pair will not have any overlapping Missing a-b segments.

For example, for siblings. Each matches half of their father and half of their mother. When comparing one with the other, they will on average not match each other over 25% of their DNA, they will match on one parent over 50% of their DNA, and they will on average match on both their parents over 25% of their DNA.

This also can apply to anyone who matches someone else through their father AND through their mother on the same segment. Cousins can fully match just as siblings do, but the double match area would cover a much smaller percentage of their DNA.

GEDmatch for example, color codes these:

image

Matches to one parent are colored yellow. Matches to both parents are colored green.

What this Corollary really means for Double Match Triangulator is that some Triangulation Groups might be triangulating to two different ancestors, one on Person a’s father’s side, and one on Person a’s mother’s side.

If there are any overlapping Missing a-b segments, then the Triangulation is guaranteed to be on one half only.

If there are not any overlapping Missing a-b segments, then the Trangulation may be over both halves.

This Corollary will provide additional help in identifying the matches that belong to each of half of the Chromosome pair.

2017 #RootsTech #InnovatorShowdown Contestants - 3 days, 14 hrs ago

I’ve been watching with much interest on the RootsTech Devpost site, some of the great programs that are being submitted to the showdown. Although the submitted programs have not yet been announced, I’ve been able to find many of them because of the videos they produced on either YouTube or Vimeo, or through their entries as participants on the Showdown site.

image

The programs that seem to be entrants for 2017 include, in alphabetical order:

  1. autodotbiography - Online website – Bryher Scudamore
  2. CSI: Crowd Sourced Indexing – Online website - Banai Lynn Feldstein
  3. The Diderot Network – Online Website – Paul Fraser
  4. Double Match Triangulator - Windows program - Louis Kessler
  5. Emberall – iOS or Android app – Karen & Kyle Corbitt, Sam Nelson
  6. Famicity - Online website - Jerome Blanchard and many others (2016)
  7. The Family Nexus - iOS app - David Taylor
  8. genealogyDOTcoach - Online website - Janet Hovorka and Kim
  9. GenealogyWallCharts – Online service – Doug Butts
  10. GenerationStory - iOS app - Shannon Uschold (2016)
  11. GenQuizitive - Online website - Melissa & John Finlay
  12. greetingStory – Online service – Christopher Cummings
  13. Irish Family History Centre - Online website – Irish Family History Centre
  14. Kindex - Online website - Cathy Gilmore (2016)
  15. Lifey – Online service - Adam Balinksi
  16. Little Family Tree - Android app - John & Melissa Finlay (2016)
  17. LongLostRelatives – Online service – John DeFour
  18. Memory Book - Online website - Chijioke Esedo, Gloria Miao and others
  19. OldNews USA - Android App - Bill Nelson
  20. Pass it Down – iOS app - Christopher Cummings & Rodger Maarfi (2016)
  21. QromaTag - iOS app - Tony Knight
  22. Rendez-Vous – Android App – Noureddine Amri
  23. RootsFinder - Online website - Heather Henderson
  24. TSOlife - Online website - David Sawyer, Stella Parris (2016)
  25. VR Tag Room360 – Online website – Tom Nguyen
  26. VyTräd - Windows program - Steven Larson

From the above, you’ll see 14 are online websites, 8 are iOS or Android apps, 2 are Windows programs, and 3 are online services, so there’s quite a variety. Many of the videos are superbly done.

6 of the programs were submitted to the 2016 Innovator Showdown. They will have gained in the experience from that and their programs will be more polished and should have a better chance this time around.

It’s going to be really tough for the judges to pick the 10 semifinalists from among these. I’m hoping DMT will impress them enough to make the first cut. There will be some good programs that won’t.

I do expect there will be a few more entrants submitted in the last few nail-biting hours left. All entries must be complete by 11:45 p.m. MST tonight.

I’ll add others to the above list as I discover them.

The following programs didn’t make the deadline:

  1. Carbon Copy – Online website – Cassidy Williams

My First YouTube Video - 5 days, 14 hrs ago

For the 2017 #RootsTech Innovator Showdown, a requirement is that the submitted program must be accompanied by a 60 to 90 second video uploaded to YouTube or Vimeo.

As of eight days ago, I had my Double Match Triangulator program all ready, I had the required project page all filled out at Devpost.com with all the screen shots set up. I just didn’t have the video and I had never done one before. And the deadline was in 10 days.

Today I finished a video that I am quite happy with and it is now part of my submission. This is it:

I had to research how to put a YouTube video together. I ran into the site compelling.tv. They had a lot of good how-to videos on how to make videos for YouTube. Their technique was to make a script of what you wanted to say and then to produce slides in PowerPoint to go along. They explain how to set up PowerPoint to produce videos for you tube.

PowerPoint! Now that would work for me. I am experienced in PowerPoint and use it for all my talks.

They then suggest getting Audacity, a free audio recording and editing software. Then set up the slides in PowerPoint to And put it together. The trick is to get the set up the timing of each slide PowerPoint to be set to display for just as long as the associated audio for the slide. That could be a bit of a nuisance, so compelling.tv sells a product called One Hour Video Studio that will create the PowerPoint file with the right timings. It actually looks like a decent system.

But I didn’t want to spend time figuring out how to use and maybe struggle with someone else’s program when I needed to make this video quickly and do it now.

I had a nice alternative. I had purchased Snagit for my home computer, mainly because I had it at work and loved it. Snagit is a screen capture program that I use to do most of my screenshots for my blog, my websites and in the help files for Behold and my other programs.

And Snagit includes basic screen recording. The nice part about it is that with Snagit, you can record the audio at the same time. That works out really nicely.

What I did first was write up my script. Then based on the script, I set up about 10 PowerPoint slides. I included fade transitions between slides as well as step by step transitions on some of the slides. In my script, I started a new paragraph each time there was a new step or slide transition.

To record, I started the PowerPoint presentation on my right monitor. I started Snagit screen recorder on my left monitor. I selected the right monitor as the area to record. I put on my headphones with its microphone and I started recording. All I had to do was read the script and then press enter when I reached the end of a paragraph to transition PowerPoint to the next step or slide. It worked well.

My first recording had a few stammers but wasn’t too bad except for one thing: the video was 2 minutes and 48 seconds long. Unfortunately, the rules stated that the video had to be between 60 and 90 seconds long. I had to cut out 78 seconds.

It took me a few hours to edit down the material and what I was saying. But that was actually worthwhile. The stuff I removed was the less important material, and I compressed some text into fewer words. About 8 takes later, I had a nice 89 second video that I liked way better than my original.

I sent it off to my friend and fellow geneablogger James Tanner who had been giving me excellent suggestions along the way on making the video. He liked what I had done. So then this morning, it took me just 4 takes to record a version without stammers that I was happy with. Voila: my first MP4 video.

I already had a YouTube account, but I had never uploaded my own video before. It wasn’t hard. Something new learned. Well worth my time.

My channel on YouTube

The deadline for the submission to the Innovator Showdown is December 1. Since I’ve now done the video, my submission now meets the requirements.

I’ve had a lot of friends wishing me good luck with this, and I thank all of you for your support. We’ll have to see first if the judges pick Double Match Triangulator to be one of the 10 semi-finalists. Those 10 get to present at the Innovator Summit Day (the Wednesday) at RootsTech, and it would be so much fun to do that. Let’s keep our fingers crossed.