Page 2 of 3 FirstFirst 123 LastLast
Results 11 to 20 of 23

Thread: Dante Labs (WGS)

  1. #11
    Registered Users
    Posts
    724
    Sex
    Location
    Glasgow, Scotland
    Ethnicity
    Pictland/Deira
    Y-DNA
    R1b-M222-FGC5864
    mtDNA
    H5r*

    Quote Originally Posted by Donwulff View Post
    New review for Dante Labs on medgadget.com
    It's not independent in that the reviewer got it for free and thus Dante Labs knew there was going to be a review, but it does look like a genuine outside review.
    I, on the other hand, sent the sample in 11 weeks ago now, with Dante Labs stating I'd receive results within couple of weeks three weeks ago, so I'm inclined to think it's a scam. As said, it wouldn't be the first time a DNA company misses delivery estimates, but given the campaign prices are so far below going price/materials price (Human whole genome sequencing at BGI starts at $600, Illumina is couple hundred more) with no plan to cover the price difference presented that they have burden of proof.
    There is also a first "Verified Purchase" review on Amazon, but it sounds & looks like Dante Labs press release. My reading is that they ordered the test on 5th September so could not actually have received the results in any case (Which is quite confusing, why admit it was ordered three weeks ago if the processing time is eight weeks?)
    Sequencing.com, their biinformatics analytics partner, has been listing Dante Labs as their "preferred provider" for a while though.
    If you read the Verified Purchase review on Amazon carefully they never actually say they have had their results yet. Its an obfuscation masterpiece. Even the phrase 'The report results are in plain English and I did not find it difficult to understand, ' is from the paragraph 'SAMPLE REPORTS'.
    YSEQ:#37; YFull: YF01405 (Y Elite 2013)
    GEDMatch: A828783 (autosomal DNA), 9427684 (GEDCOM) for segment matching DO NOT POST ADMIXTURE REPORTS USING MY KIT
    WGS (Full Genomes Nov 2015)
    Ancestry GCs: Scots in central Scotland & Ulster, Ireland; English in Yorkshire & Pennines
    Hidden Content

  2. #12
    Junior Member
    Posts
    8
    Sex

    I actually received my results yesterday. That makes it 79 days, or 11 weeks and two days, since they confirmed receiving the sample, as I said missing the delivery estimate is pretty common... BAM file was not included, but since their web-site says it is available (After I asked about it from them), I have requested for it, we'll see how it goes. As for the VCF file they make available, it's near perfect match with my 23andMeV3 results, and looks like a perfect half-match to one of my parents exome-sequence. This proves beyond shadow of doubt that it's an actual sequence, and not just imputed SNP-microarray/chip result, which was my first concern.
    For the casual user, the "Wellness and Longevity" report is the main feature. It's a little underwhelming, for me at least, though as indicated I've seen my own 23andMe and Promethease results as well as one parent's exome analysis, and these reports only cover SNP's which are generally tested by 23andMe already. The traits included are Athletic Performance, Melanoma, Arthritis, Osteoporosis, Malignant Hyperthermia, Heart Attack, Medication Assasment (Warfarin, Clopidogrel/Plavix, Aspirin, Statins, Blood Clot Risk including Deep Vein Thrombosis), Preventable Sudden Death (due to a Heart Arrythmia) and finally Genetic Profile which increases one-liners for Atrial Fibrillation, MTHFR Deficiency, Salt-Sensitive Hypertension, Effect of Breastfeeding as a Baby on IQ, Congenital Bilateral Absence of Vas Deferens, Obstructive Azoospermia, Lactose Intolerance, Liver Disease, Hirschsprung Disease, Blood Clot Risk (due to Factor V Leiden, Prothrombin 20210, Protein C Deficiency, Protein S Deficiency, Antithrombin III Deficiency), Bleeding Risk (due to Factor V Deficiency, Factor XI Deficiency, Von Willebrand Disease Type 2M, Prothrombin Deficiency, Von Willebrand Disease Type 1, Von Willebrand Disease Type 2, Hemochromatosis, Drug-induced Hemolysis, Hemolytic Anemia due to Triosephosphate Isomerase Deficiency, Homocysteinuria, Thrmbotic Thromboctopenic Purpurra, Idiopathic Pancreatitis, Sysceptibility to Lead Poisoning due to ALAD Gene, Brain Aneurysm, Alpha-1-Antirypsin Deficiency, Asthma, Bronchiectasis, Ephysema, Chronic Respiratory Disease, Resistance to HIV Infection, Susceptibility to West Nile Virus, Recurrent Bacterial Infections, Increased Risk of Disseminated Infection with Mycobacterium avium and Salmonella enteritidis and Noise-inducd Hearing Loss.
    Phew, that's a LOT. My main concern is that you get access to that report without a single click-wrap warning. Some people won't handle well learning that they have "Heart Attack" or "Sudden Death", for example, with no forced warning or disclaimer about how the results are only according to knowledge now, may contain errors, and in any case only a risk. 23andMe had multiple click-wraps and disclaimers but still got into hot water, which is why Dante Labs is only available in Europe, but for ethical reasons I hope they will introduce clear warnings both before order and before viewing results. If you are taking the test for genealogical purposes only, consider carefully the possible impact on yourself *and your family* if you're found positive for any of the risks above before clicking on "Wellness and Longevity" in the kit manager.
    Finally, there is a "Genome Overview", which is an SnpEff report for the VCF file. Most of the information in it, I daresay, holds little relevance or usability, what am I supposed to compare the quality metrics etc. on here? It shows there were 3,431 variants detected on Y chromosome, though. The report shows mtDNA incorrect, and the reference sequence in the VCF looks to actually be UCSC hg19 chrM, which isn't compatible with anything, really. So the BAM will be essential for phylogenetic uses.

  3. The Following 4 Users Say Thank You to Donwulff For This Useful Post:

     karwiso (10-06-2017), MacUalraig (10-09-2017), Mich Glitch (10-08-2017), sktibo (10-06-2017)

  4. #13
    Junior Member
    Posts
    9
    Sex
    Location
    Stockholm, Sweden
    Ethnicity
    Ingrian + Russian
    Nationality
    Swedish, Russian

    Finland Russian Federation Sweden
    I am testing for genealogical purposes so I am interested to upload WGS to GEDmatch.com. When I clilck on upload VCF to Genesis I see "Copy the FILE URL from your testing company, and paste into this box. Your data must be Build 37 (GRCh37/hg19) " Probably it is not that bad with hg19? FTDNA just announced the switch to GRCh38.

  5. #14
    Junior Member
    Posts
    8
    Sex

    Dante Labs was currently on UCSC hg19. It's just that the mitochondrial reference in that is way outdated.
    FTDNA's switch is for Big Y's Y-chromosome only, it won't matter for this as YFull would redo the mapping to a specific build anyway.

    Dante Labs provided a download link for the VCF file, which worked as-is for both GEDMatch Genesis and SNPedia Promethease. There's a general problem with the sequencing services though, due to the large amount of data the VCF's only list variants that differ from reference. GEDMatch Genesis will therefore take only those SNP's into account in matching, which means you're working with very limited (and skewed) set of SNP's. This is a non-trivial problem because even a whole genome sequence doesn't get sufficient coverage across the whole genome, so you can't just assume non-listed variants are reference allele either. The solution is for everybody to start using gVCF (And, preferably, some interchangeable common format...) which lists regions that match the reference.

    I have no idea if GEDMatch or SNPedia support gVCF yet, I haven't seen a statement for that. It's possible to expand out the gVCF REF-blocks, but Dante Labs VCF doesn't contan them currently, and expanding them would hugely increase the file size (Storage + file-transfer). Another quick-and-dirty fix would be to run the VCF through an imputation process, which would fill out the reference alleles, and hopefully impute some of the variants in the non-tested regions. However, currently I'm not sure there ARE any public imputation services that would take a sequencing VCF.

    I was offered the BAM file on a mailed USB-stick, but given it would take a week or two to upload the BAM file on Dropbox/Google Drive etc. (Oh, and need expanded capacity too) over a fast consumer internet connection, I asked if they could arrange for a download link, still waiting for a response. YFull does not currently list Dante Labs as a testing company on their order form, but given they frequently add samples from studies with different workflows, I'm sure they can add Dante Labs BAM files easily enough.

  6. #15
    Junior Member
    Posts
    8
    Sex

    Whoops double post deleted.
    Last edited by Donwulff; 10-09-2017 at 05:17 PM.

  7. #16
    Junior Member
    Posts
    9
    Sex
    Location
    Stockholm, Sweden
    Ethnicity
    Ingrian + Russian
    Nationality
    Swedish, Russian

    Finland Russian Federation Sweden
    I would like to provide a link, but I am not allowed to post links here yet. Please google: NCBI Genome Remapping Service NCBI
    It is free remapping tool and you can choose genome build and input and output file types.

    Well, I hope they can provide download links. I have ordered several tests and would be disappointing to go through snail mail with USB sticks...

  8. #17
    Thank you so much for updating Donwulff! I was getting worried that I'd been scammed. They received my kit 9 and a half weeks ago. I contacted them late last week and they said I could expect to get my results October 20th.
    Last edited by rbtlsr; 10-09-2017 at 07:52 PM.

  9. #18
    Registered Users
    Posts
    316
    Sex
    Location
    United States
    Nationality
    US
    Y-DNA
    R1b U106 Z326 Z81
    mtDNA
    K1a1b1a

    Quote Originally Posted by Donwulff View Post
    I actually received my results yesterday. That makes it 79 days, or 11 weeks and two days, since they confirmed receiving the sample, as I said missing the delivery estimate is pretty common... BAM file was not included, but since their web-site says it is available (After I asked about it from them), I have requested for it, we'll see how it goes. As for the VCF file they make available, it's near perfect match with my 23andMeV3 results, and looks like a perfect half-match to one of my parents exome-sequence. This proves beyond shadow of doubt that it's an actual sequence, and not just imputed SNP-microarray/chip result, which was my first concern.
    For the casual user, the "Wellness and Longevity" report is the main feature. It's a little underwhelming, for me at least, though as indicated I've seen my own 23andMe and Promethease results as well as one parent's exome analysis, and these reports only cover SNP's which are generally tested by 23andMe already. The traits included are Athletic Performance, Melanoma, Arthritis, Osteoporosis, Malignant Hyperthermia, Heart Attack, Medication Assasment (Warfarin, Clopidogrel/Plavix, Aspirin, Statins, Blood Clot Risk including Deep Vein Thrombosis), Preventable Sudden Death (due to a Heart Arrythmia) and finally Genetic Profile which increases one-liners for Atrial Fibrillation, MTHFR Deficiency, Salt-Sensitive Hypertension, Effect of Breastfeeding as a Baby on IQ, Congenital Bilateral Absence of Vas Deferens, Obstructive Azoospermia, Lactose Intolerance, Liver Disease, Hirschsprung Disease, Blood Clot Risk (due to Factor V Leiden, Prothrombin 20210, Protein C Deficiency, Protein S Deficiency, Antithrombin III Deficiency), Bleeding Risk (due to Factor V Deficiency, Factor XI Deficiency, Von Willebrand Disease Type 2M, Prothrombin Deficiency, Von Willebrand Disease Type 1, Von Willebrand Disease Type 2, Hemochromatosis, Drug-induced Hemolysis, Hemolytic Anemia due to Triosephosphate Isomerase Deficiency, Homocysteinuria, Thrmbotic Thromboctopenic Purpurra, Idiopathic Pancreatitis, Sysceptibility to Lead Poisoning due to ALAD Gene, Brain Aneurysm, Alpha-1-Antirypsin Deficiency, Asthma, Bronchiectasis, Ephysema, Chronic Respiratory Disease, Resistance to HIV Infection, Susceptibility to West Nile Virus, Recurrent Bacterial Infections, Increased Risk of Disseminated Infection with Mycobacterium avium and Salmonella enteritidis and Noise-inducd Hearing Loss.
    Phew, that's a LOT. My main concern is that you get access to that report without a single click-wrap warning. Some people won't handle well learning that they have "Heart Attack" or "Sudden Death", for example, with no forced warning or disclaimer about how the results are only according to knowledge now, may contain errors, and in any case only a risk. 23andMe had multiple click-wraps and disclaimers but still got into hot water, which is why Dante Labs is only available in Europe, but for ethical reasons I hope they will introduce clear warnings both before order and before viewing results. If you are taking the test for genealogical purposes only, consider carefully the possible impact on yourself *and your family* if you're found positive for any of the risks above before clicking on "Wellness and Longevity" in the kit manager.
    Finally, there is a "Genome Overview", which is an SnpEff report for the VCF file. Most of the information in it, I daresay, holds little relevance or usability, what am I supposed to compare the quality metrics etc. on here? It shows there were 3,431 variants detected on Y chromosome, though. The report shows mtDNA incorrect, and the reference sequence in the VCF looks to actually be UCSC hg19 chrM, which isn't compatible with anything, really. So the BAM will be essential for phylogenetic uses.
    Thus far, Promethease remains superior to most commercial DTC health reports.

  10. #19
    Registered Users
    Posts
    724
    Sex
    Location
    Glasgow, Scotland
    Ethnicity
    Pictland/Deira
    Y-DNA
    R1b-M222-FGC5864
    mtDNA
    H5r*

    Quote Originally Posted by Donwulff View Post
    Dante Labs was currently on UCSC hg19. It's just that the mitochondrial reference in that is way outdated.
    FTDNA's switch is for Big Y's Y-chromosome only, it won't matter for this as YFull would redo the mapping to a specific build anyway.

    Dante Labs provided a download link for the VCF file, which worked as-is for both GEDMatch Genesis and SNPedia Promethease. There's a general problem with the sequencing services though, due to the large amount of data the VCF's only list variants that differ from reference. GEDMatch Genesis will therefore take only those SNP's into account in matching, which means you're working with very limited (and skewed) set of SNP's. This is a non-trivial problem because even a whole genome sequence doesn't get sufficient coverage across the whole genome, so you can't just assume non-listed variants are reference allele either. The solution is for everybody to start using gVCF (And, preferably, some interchangeable common format...) which lists regions that match the reference.

    I have no idea if GEDMatch or SNPedia support gVCF yet, I haven't seen a statement for that. It's possible to expand out the gVCF REF-blocks, but Dante Labs VCF doesn't contan them currently, and expanding them would hugely increase the file size (Storage + file-transfer). Another quick-and-dirty fix would be to run the VCF through an imputation process, which would fill out the reference alleles, and hopefully impute some of the variants in the non-tested regions. However, currently I'm not sure there ARE any public imputation services that would take a sequencing VCF.

    I was offered the BAM file on a mailed USB-stick, but given it would take a week or two to upload the BAM file on Dropbox/Google Drive etc. (Oh, and need expanded capacity too) over a fast consumer internet connection, I asked if they could arrange for a download link, still waiting for a response. YFull does not currently list Dante Labs as a testing company on their order form, but given they frequently add samples from studies with different workflows, I'm sure they can add Dante Labs BAM files easily enough.
    My WGS FGC reports include a dbSNP.vcf report of about 41Gb which includes all calls ref or alt for SNPs that occur in the dbSNP database. The other main VCF is a snpeff one (2 Gb) which is restricted to non-ref results.
    YSEQ:#37; YFull: YF01405 (Y Elite 2013)
    GEDMatch: A828783 (autosomal DNA), 9427684 (GEDCOM) for segment matching DO NOT POST ADMIXTURE REPORTS USING MY KIT
    WGS (Full Genomes Nov 2015)
    Ancestry GCs: Scots in central Scotland & Ulster, Ireland; English in Yorkshire & Pennines
    Hidden Content

  11. #20
    Junior Member
    Posts
    8
    Sex

    About a week after I requested BAM & gave an impassioned plea for download link, they sent it on a 128GB SanDisk Cruzer USB-stick via DHL. Good news is it was an overnight delivery, but this seems untenable for most customers without data-processing tools at home and opens some new privacy concerns for people who believe in genetic exceptionalism (Ie. that the information couldn't be obtained in other ways). For myself, I have mobile broadband on a tablet, and made it a WLAN access point for my Windows laptop.

    Uploading the ~100 gigs BAM to Amazon S3, I first tried Cloudberry Explorer which offers 14 day Pro trial with multipart upload. This was horrible; with current version, the percent complete calculation was totally off (And actually showed two values), the time left showed the expected total transfer time, and most importantly, when the upload finished 75 blocks failed. I could find no way to either resume the download or even delete the unfinished file so I wouldn't have to pay storage fee for it. S3 Browser worked nice enough, letting me view and delete incomplete multipart files, and again most importantly, finish the upload without errors even though I suspended the laptop without warning at one point.

    I'm expecting Amazon AWS storage and transfer fees to be in the neighborhood of 5 euros, maybe less given I only need to store it until YFull downloads it. Considering the USB stick is 30 dollars and DHL shipping probably in the neighborhood of 50 dollars, this is obviously un-optimal, though if Dante Labs does their processing offline they could be facing similar upload bandwidth problem. Nonetheless, I hope they solve that with further sequences. Having ordered my sequence at their Amazon Prime Day discount, it also seems like just the DHL, sample processing and USB costs would put them in the red, never mind the actual sequencing; it's likely they're burning investor money to get established, but I'm still not getting their business model exactly.

    YFull replied they think they can process the BAM file, and there doesn't seem to be anything out of ordinary with it, but will probably be a while before I know for sure.

    As an aside, I could impute the 166MB VCF-file that Dante Labs provides standard with Sanger Imputation Service (HRC 1.1), but this isn't currently an easy task. The service can't handle chrM or chrY, and the chromosomes need to be renamed to GRCh37 without the "chr" part, and Globus Connect is needed for transferring the files. After that you still need to combine the chromosomes into a single file that's close to the "dbSNP" list (with potentially unsequenced variants imputed and ref-calls filled out) and do whatever preparation your intended use requires. I could not yet get GedMatch to accept the imputed file, and it provides no error so I don't know if it's file size or what.

  12. The Following User Says Thank You to Donwulff For This Useful Post:

     karwiso (10-13-2017)

Page 2 of 3 FirstFirst 123 LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •