Degui Zhi

@zhizhid

Associate Professor . Develop methods connecting humans via Identity-by-Descent. Tune in to Deep Learning, AI, EHR, Genomics.

Vrijeme pridruživanja: travanj 2009.

Tweetovi

Blokirali ste korisnika/cu @zhizhid

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

  1. proslijedio/la je Tweet
    29. sij

    Medicine is going to be changed by the application of ML to health care and biomedical discovery. DBMI is looking for AI/ML researchers to join our tenure-track faculty to lead research to effect that change in the most effective, safe and rapid way.

    Poništi
  2. 28. sij
    Poništi
  3. 26. sij

    In terms of compute per watts GPUs still have a long way to go to catch up BPUs (bird processing units).

    Poništi
  4. 26. sij

    Finally the speed of genomics is catching up with that of Deep learning.

    Poništi
  5. proslijedio/la je Tweet
    24. sij

    Open access statistical population genomics textbook with example code.

    Poništi
  6. proslijedio/la je Tweet

    Choudhary, Wang and co look at how transposable elements contribute to genome 3D structure. About a fifth of CTCF loop anchors in mouse and human are contributed by repetitive elements. TEs contribute to turnover of CTCF binding sites.

    Poništi
  7. proslijedio/la je Tweet
    22. sij
    Poništi
  8. proslijedio/la je Tweet
    23. sij

    Thanks to open data sharing in GISAID by the Guangdong Provincial Center for Diseases Control, we've updated with 6 new genomes from Guangdong. These show clear evidence of human-to-human transmission with 2 clusters of related viruses.

    Prikaži ovu nit
    Poništi
  9. proslijedio/la je Tweet
    18. sij
    Poništi
  10. proslijedio/la je Tweet
    16. sij

    Machine Learning Hackathon for Cancer Drug Discovery is now open for registration. Organized by UTHealth School of Biomedical Informatics. February 1-2, 2020. Unlimited coffee and food and $1200 prize.

    Poništi
  11. proslijedio/la je Tweet
    16. sij

    This is potentially very useful, allowing the PBWT data structure to change dynamically with continuous data collection.

    Poništi
  12. proslijedio/la je Tweet
    16. sij

    d-PBWT: dynamic positional Burrows-Wheeler transform

    Poništi
  13. 16. sij

    Excited to share our newest preprint from the young and talented Ahsan Sanaullah and . We introduced the dynamic version of the beloved PBWT. Accepted to .

    Poništi
  14. proslijedio/la je Tweet
    7. sij

    Two studies suggest that a determined adversary may be able to obtain genetic information without permission from some genealogy databases

    Quote by Shai Carmi - "Therefore, I urge all stakeholders to pay attention to the work of these two groups and attempt to keep genetic information secure"
    Poništi
  15. proslijedio/la je Tweet
    7. sij

    Measuring genetic variation in the multi-ethnic Million Veteran Program (MVP)

    Poništi
  16. proslijedio/la je Tweet
    8. sij
    Poništi
  17. proslijedio/la je Tweet

    New study by and colleagues introduces CaSpER: a statistical framework that uses RNA-sequencing reads to identify and visualise copy number variants events @ UTHealth_SBMI

    Poništi
  18. proslijedio/la je Tweet
    2. sij

    please RT: looking for a to work on large scale detection. We will analyze multiple thousands of genomes + develop methods to identify novel targets for diseases: , ..

    Prikaži ovu nit
    Poništi
  19. 28. pro 2019.

    My student Laila Rasmy starts at -1:22!

    Poništi
  20. proslijedio/la je Tweet
    24. pro 2019.

    NEW from Hongyang Li and colleagues at : Joint learning improves protein abundance prediction in Read more here:

    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:

    ·