Andrew Carroll

@acarroll_ATG

Product lead, Genomics . All about science, machine learning, bioinformatics, and genomics. Opinions are my own. he/him

Vrijeme pridruživanja: travanj 2015.

Tweetovi

Blokirali ste korisnika/cu @acarroll_ATG

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @acarroll_ATG

  1. proslijedio/la je Tweet
    prije 7 sati

    Looking for a new opportunity at a high-growth biotechnology startup bringing AI/ML to drug discovery? BigHat is recruiting a Technology Development Specialist to build our high-throughput, automated platform for antibody production and characterization

    Poništi
  2. 30. sij

    I really like this approach. I had a similar reaction about mapping after reading the Learned Index Structure paper. This work looks interesting and promising. Relevant to mapping, use of minimizers, similarity detection, and maybe assembly. Very cool.

    Poništi
  3. proslijedio/la je Tweet
    27. sij

    HG002-DataFreeze 1.0 is now available to help guide new assembly and evaluation methods. Data sharing is in partnership with and .

    Poništi
  4. 21. sij
    Poništi
  5. proslijedio/la je Tweet
    15. sij

    Don't miss our update on Princess developer workshop starting in 30 min (Golden West Room). Together with many other exciting talks from , etc. Hope to see many of you there

    Poništi
  6. 3. sij

    "If you missed the opportunity to participate in the first round and you think your project fits with the goals of this program, please consider applying. Applications for the second cycle open on December 16, 2019, and close February 4, 2020."

    Poništi
  7. proslijedio/la je Tweet

    Want to help evaluate the new v4.1 small variant benchmark for RM8391 (HG002) that enables you to assess accuracy in more difficult regions, including many segmental duplications, difficult-to-map regions, and the MHC? See details at

    Poništi
  8. proslijedio/la je Tweet
    22. pro 2019.

    What's striking to me about this is the top reasons: - HR says: Women don't receive as much sponsorship - Men say: There are fewer qualified women in the pipeline - Women say: Women are judged by different standards

    Poništi
  9. 20. pro 2019.

    If you are coming to , stay for the Jan15 meeting (). I'll talk on variant calling with HiFi data, incl difficult regions, and results from new projects. Other great speakers

    Poništi
  10. proslijedio/la je Tweet

    ML performs best with large datasets, but sometimes there isn't as much data! In this blog post from our genomics team, Ren Yi investigates how to best learn from data in a larger domain to improve on a specialized one. 🧬 💡Learn more →

    Genomics with TensorFlow
    Poništi
  11. 6. pro 2019.

    For a deeper technical dive, please see the NeurIPS submission "Learning from Data Rich Problems" . Ren's intern mentors and  were also instrumental. Ren and  will be at NeurIPS ML4H, please find them and talk science & ML!

    Prikaži ovu nit
    Poništi
  12. 6. pro 2019.

    This blog details work from Ren Yi in her summer internship. It investigates strategies to improve ML performance in specialized problems (in this, exomes) by using similar, data-rich areas (WGS). Ren will present this work at ML4H workshop

    Prikaži ovu nit
    Poništi
  13. proslijedio/la je Tweet
    5. pro 2019.

    Clair is the successor of Clairovoyante. For ONT data, Clair achieves the best precision, recall, and speed as compared to several competing programs, including Clairvoyante, Longshot, and Medaka.

    Poništi
  14. 5. pro 2019.

    Very nice method and investigation. There are many good investigations about training (e.g. use of downsampling to improve general performance and generalization across references).

    Poništi
  15. proslijedio/la je Tweet
    4. pro 2019.

    Nice Nature article on visualization in genomics! Some nice discussion from of disseminating results through creating new browser tracks, and even a few nuggets at the end from yours truly 😅

    Prikaži ovu nit
    Poništi
  16. proslijedio/la je Tweet
    3. pro 2019.

    Very excited to announce my latest WebAssembly project: 🧬 It's a playground for samtools, a genomics CLI that I ported from C to so it can run directly in the browser!

    Prikaži ovu nit
    Poništi
  17. proslijedio/la je Tweet
    22. srp 2019.

    Oh look, it's samtools running in the browser! 😲 This is just a preview, stay tuned for more.

    Poništi
  18. 25. stu 2019.

    This is fantastic work! One thing I haven't seen mentioned yet is the transmission rate in Table S1 for HET+HET parents -> (72% HET). A summary of events violating HWE could be very valuable for refining methods (along with a run of HG002)

    Poništi
  19. 14. stu 2019.

    It will be really interesting to see how this develops as a model for supporting bioinformatics methods goes. I could see this as a template for a pre-competitive initiative from industry too. Congratulations to the funded proposals!

    Poništi
  20. proslijedio/la je Tweet
    13. stu 2019.

    If you're calling small variants with , check out the newest version of DeepVariant, v0.9, with a new model trained including PacBio Sequel II System Chemistry 2.0 data. Internal testing shows improved precision and recall for all previous chemistries.

    Poništi

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·