Nebula Genomics

[Ibericus]

I logged in to see how to register my kit and I saw that their lifetime membership is $499 again. I have $50 coupon that I got when I ordered two weeks ago. It may be expired, I don't know. PM if interested.

[dosas]

Please check your inbox.

Can you forward the message to me, please?

I am waiting on my Nebula 30x, it won't be long now I think, and I would like to generate a super-combined RAW kit, as we all as the 23ame v3, you mentioned.

[MagicalRaindrops]

Please check your inbox.

Could you send the instructions to me as well? I haven't found anything clear on google.

[aaronbee2010]

Could you send the instructions to me as well? I haven't found anything clear on google.

Since you're a new user, you won't be able to send me private messages until you reach 15 posts on this forum, so I'll keep things on here for now. Also, before I go on any further, I need to ask you whether you have a 0.4x BAM or 30x CRAM from Nebula as the instructions you need to follow are dependant on that.

[MagicalRaindrops]

Oh, alright. I have the 30x CRAM.

[aaronbee2010]

Oh, alright. I have the 30x CRAM.

Okay, so you'll need to convert the CRAM file to a BAM file before you can proceed with the generation of the microarray file.

What operating system are you using, what CPU does your computer have, how much RAM does your computer have and how much free space do you have left on your computer?

You will need a CPU with at least 4 threads, preferably at least a Core i5 (or a newer Core i3) or Ryzen 3, at least 6 GB RAM (but preferably at least 8 GB) and at least 200 GB of free space remaining on your computer (at least on the hard drive you're downloading to) but preferably at least 250 GB.

[MagicalRaindrops]

Okay, so you'll need to convert the CRAM file to a BAM file before you can proceed with the generation of the microarray file.

What operating system are you using, what CPU does your computer have, how much RAM does your computer have and how much free space do you have left on your computer?

You will need a CPU with at least 4 threads, preferably at least a Core i5 (or a newer Core i3) or Ryzen 3, at least 6 GB RAM (but preferably at least 8 G and at least 200 GB of free space remaining on your computer (at least on the hard drive you're downloading to) but preferably at least 250 GB.

Win 10, Ryzen 7 2700X, 32 GB, and 500+ GB.

[aaronbee2010]

Win 10, Ryzen 7 2700X, 32 GB, and 500+ GB.

That's quite the beast you've got there. Since you have 16 threads at your disposal, I'll write the instructions below using commands set to use 12 threads.

Firstly, you'll need to download WGSExtract (Program: http://37.187.22.93/wgsextract/WGSExtractBeta.zip - Manual: http://bit.ly/36Jdpnq). When downloaded, right-click WGSExtractBeta.zip and click on "Extract all" then click "Extract" without changing anything. The following instructions assume that you've downloaded WGSExtractBeta.zip to the same folder as your CRAM and CRAI files and that you used the default extraction destination provided after clicking "Extract all"

Load the command prompt (you don't need to run it as administrator), move to the folder where you downloaded WGSExtract and your Nebula data, then copy the following into there:

WGSExtractBeta\WGSExtract\programs\samtools-mingw\samtools.exe view [email protected] -hb -T WGSExtractBeta\WGSExtract\reference_genomes\GCA_00 0001405.15_GRCh38_no_alt_analysis_set.fna.gz -o X.bam X.cram && WGSExtractBeta\WGSExtract\programs\samtools-mingw\samtools.exe index [email protected] X.bam

Where X.cram is your CRAM file and X.bam is the BAM file being produced from it.

Once this is done, you should hopefully have a BAM file and it's corresponding BAI index. You can then go to WGSExtractBeta\WGSExtract and click "Windows_START.bat". After WGSExtract has loaded, choose your language, select the BAM file and the output folder then go to the "Extract data" tab, click "Generate files in several autosomal formats", select what formats you want to generate. I recommend selecting the "Combined file with all SNPs from the formats of all DNA test providers (RECOMMENDED)" and "23andme v3 (11/2010 - 11/2013)" options together. The former gets the highest SNP count on GEDmatch, Admixture Studio, GenoPlot, Global25 etc. and the latter has the best compatibilty for uploading to sites like FTDNA, MyHeritage and LivingDNA, to name a few examples. Pretty much all chip vendors and versions are available, in case you have a specific one in mind.

When this starts running, you should go to Task Manager and make sure there's a Python process running with noticeable CPU usage (I usually get around 15-17% CPU usage from Python when WGSExtract is working - I have a i7-4770K @ 4.0 GHz for reference). If this isn't the case, try restarting WGSExtract and trying again until it works. From my experience, I had to repeat this several times on one day, then on another day it would work first time.

When it's working, it will probably take up to a few hours until it's done. The notification window will remain frozen until the process is complete. You'll get a notification saying "Your BAM file was aligned to the reference genome hg38. The resulting autosomal files will be less accurate than those using a hs37d5 BAM file as source.". The difference is insignificant though. For me, the combined microarray file from my hs37d5 BAM gave me 188163/188173 (-10) positions used on HarappaWorld using Admixture Studio and the combined microarray file from my hg38 BAM gave me 188132/188173 (-41) positions used. For reference, my MyHeritage v1 file gets 181936/188173 (-6237) positions used and my 23andMe v5 file gets 52646/188173 (-135527) positions used, so I wouldn't worry much about using a hg38 file. Since Nebula only provide a CRAM file in hg38, you would have to generate a hs37d5 BAM from the FASTQ files, which is a time-consuming process. For reference, It takes me around 3 days to produce a hs37d5 BAM from 30x FASTQ data using a GATK-based workflow, which is considered one of the gold standards in the bioinformatics industry, and I already have all the required programs and files set up.

After that's complete, you can go to "Generate BAM file that contains both Y and mtDNA" to extract Y-DNA and mtDNA data from the BAM file to a seperate Y+mtDNA BAM file which can be uploaded to YFull for Y-DNA and mtDNA analysis if you want to. There are other functions available in WGSExtract, which you're free to peruse yourself.

I hope all of this helps you. Please let me know if there are any issues.

[MagicalRaindrops]

That's quite the beast you've got there. Since you have 16 threads at your disposal, I'll write the instructions below using commands set to use 12 threads.

Firstly, you'll need to download WGSExtract (Program: http://37.187.22.93/wgsextract/WGSExtractBeta.zip - Manual: http://bit.ly/36Jdpnq). When downloaded, right-click WGSExtractBeta.zip and click on "Extract all" then click "Extract" without changing anything. The following instructions assume that you've downloaded WGSExtractBeta.zip to the same folder as your CRAM and CRAI files and that you used the default extraction destination provided after clicking "Extract all"

Load the command prompt (you don't need to run it as administrator), move to the folder where you downloaded WGSExtract and your Nebula data, then copy the following into there:



Where X.cram is your CRAM file and X.bam is the BAM file being produced from it.

On my screen, it looks like the reference name has a space within it (GCA_000001405.15_GRCh 38_no_alt_analysis_set.fna.gz) between GRCh and 38 when I typed it out, but there's supposed to be no space. Just mentioning this in case you have the same issue as I do.

Once this is done, you should hopefully have a BAM file and it's corresponding BAI index. You can then go to WGSExtractBeta\WGSExtract and click "Windows_START.bat". After WGSExtract has loaded, choose your language, select the BAM file and the output folder then go to the "Extract data" tab, click "Generate files in several autosomal formats", select what formats you want to generate. I recommend selecting the "Combined file with all SNPs from the formats of all DNA test providers (RECOMMENDED)" and "23andme v3 (11/2010 - 11/2013)" options together. The former gets the highest SNP count on GEDmatch, Admixture Studio, GenoPlot, Global25 etc. and the latter has the best compatibilty for uploading to sites like FTDNA, MyHeritage and LivingDNA, to name a few examples. Pretty much all chip vendors and versions are available, in case you have a specific one in mind.

When this starts running, you should go to Task Manager and make sure there's a Python process running with noticeable CPU usage (I usually get around 15-17% CPU usage from Python when WGSExtract is working - I have a i7-4770K @ 4.0 GHz for reference). If this isn't the case, try restarting WGSExtract and trying again until it works. From my experience, I had to repeat this several times on one day, then on another day it would work first time.

When it's working, it will probably take up to a few hours until it's done. The notification window will remain frozen until the process is complete. You'll get a notification saying "Your BAM file was aligned to the reference genome hg38. The resulting autosomal files will be less accurate than those using a hs37d5 BAM file as source.". The difference is insignificant though. For me, the combined microarray file from my hs37d5 BAM gave me 188163/188173 (-10) positions used on HarappaWorld using Admixture Studio and the combined microarray file from my hg38 BAM gave me 188132/188173 (-41) positions used. For reference, my MyHeritage v1 file gets 181936/188173 (-6237) positions used and my 23andMe v5 file gets 52646/188173 (-135527) positions used, so I wouldn't worry much about using a hg38 file. Since Nebula only provide a CRAM file in hg38, you would have to generate a hs37d5 BAM from the FASTQ files, which is a time-consuming process. For reference, It takes me around 3 days to produce a hs37d5 BAM from 30x FASTQ data using a GATK-based workflow, which is considered one of the gold standards in the bioinformatics industry, and I already have all the required programs and files set up.

After that's complete, you can go to "Generate BAM file that contains both Y and mtDNA" to extract Y-DNA and mtDNA data from the BAM file to a seperate Y+mtDNA BAM file which can be uploaded to YFull for Y-DNA and mtDNA analysis if you want to. There are other functions available in WGSExtract, which you're free to peruse yourself.

I hope all of this helps you. Please let me know if there are any issues.

Hmm, I've done some of that, but not that specific cram to bam method. I'll try that and see what comes of it.

Later note: I had no idea that samtools supported multiple threads, so thanks for pointing that out!

[sktibo]

Just checking in - IIRC nebula was planning on including an ftdna myorigins estimate with their test? Hopefully that's in the works still? (along with the now long awaited update to myorigins itself..)