Dylan Marshall

@dyl4nm4rsh4ll

ML / genomics

Cambridge, MA
Vrijeme pridruživanja: rujan 2016.

Tweetovi

Blokirali ste korisnika/cu @dyl4nm4rsh4ll

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @dyl4nm4rsh4ll

  1. proslijedio/la je Tweet
    18. pro 2019.

    Hi friends, I've started a lab at Harvard and I'm looking to hire a post-doc! We identify genetic associations in natural populations, test them by making mutants in the lab, and study the development that brings genotype and phenotype together = ). Ad:

    Prikaži ovu nit
    Poništi
  2. proslijedio/la je Tweet
    21. sij
    Odgovor korisnicima

    If you are covering pseudolikelihood, might be worth including the original paper that introduced pll for contact prediction and protein design:

    Poništi
  3. proslijedio/la je Tweet
    6. sij

    We're looking for a PhD-level research scientist! Come help us elucidate how gene expression processes are coordinated, from the nucleus to the mitochondria. If you are considering a non-PI career path, but want to stay in academic research, please apply!

    Prikaži ovu nit
    Poništi
  4. 15. pro 2019.

    Status: currently overwhelmed by ideas from . Check out poster / code for our work on understanding what generative models learn from protein sequences 😎 👇🏼

    Poništi
  5. proslijedio/la je Tweet

    4. What do generative models learn from protein sequences? Dylan Marshall Haobo Wang

    Prikaži ovu nit
    Poništi
  6. 8. pro 2019.

    Am excited to try this out! Another exotic neural network visualization method from the . Lucidly explained by .

    Poništi
  7. proslijedio/la je Tweet
    28. stu 2019.

    Rock climbing is my therapy: -being outdoors reminds me there are some things that nobody can take away from me -focusing intently on a boulder problem makes everything else fade into the background -tackling tough challenges reminds me just how strong I am What’s your therapy?

    Poništi
  8. proslijedio/la je Tweet

    When an deep learning model backfires: —High accuracy but no performance improvement —Incorrect model predictions strongly biased doctor's diagnosis —Very instructive

    Poništi
  9. proslijedio/la je Tweet
    27. lis 2019.
    Prikaži ovu nit
    Poništi
  10. proslijedio/la je Tweet
    27. ruj 2019.

    The GeneWalk tutorial is online on the website. It explains install / run of GW, its methodology, output and data analysis with on how to plot gene function relevance score bar charts in and .

    Poništi
  11. proslijedio/la je Tweet
    12. ruj 2019.

    I am excited to announce that I have started my new lab in the Simons Center for Quantitative Biology at Cold Spring Harbor Laboratory and we have open postdoc positions! For more info:

    Poništi
  12. 5. ruj 2019.

    Question: how to discern salient changes in and contextual functionality of gene expression between wild type and perturbed environments? Answer: 👇 ... congrats !

    Poništi
  13. 23. kol 2019.

    Outrageous bullying, academia’s dirty secret: ‘Paralysed by anxiety’: researchers speak about life in troubled ancient-DNA lab

    Poništi
  14. 19. svi 2019.

    "The cloud", what a concept from . One of the more romantic descriptions of how science takes place. Recommended video: .

    Poništi
  15. proslijedio/la je Tweet
    17. tra 2019.
    Prikaži ovu nit
    Poništi
  16. proslijedio/la je Tweet
    18. velj 2019.
    Poništi
  17. 12. velj 2019.
    Poništi
  18. 2. pro 2018.
    Poništi
  19. 26. lis 2018.

    ~ 1 / 4 bioinformatics tools “impossible to install” 😂 is anyone surprised?

    Poništi
  20. 14. lis 2018.

    Me reading my cousin’s latest math paper “Approximating mixed Hölder functions using random samples” -

    Prikaži ovu nit
    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:

    ·