James Hawley @jrhawley@scholar.social
- Website
- https://jrhawley.github.io
- GitHub
- https://github.com/jrhawley
- Personal Mastodon
- https://toot.cafe/@jrhawley
PhD Candidate at the University of Toronto, in the Department of Medical Biophysics.
I use my math and physics background to study computational biology in epigenetics and cancer in laboratory of Dr. Mathieu Lupien.
This is my scientific account. Views are my own.
Joined Feb 2019
Pinned toot
I'm a PhD candidate in Medical Biophysics at the University of Toronto studying epigenetics and cancer. Math and physics background taking a step into bioinformatics and evolution #introductions
James Hawley
boosted
For example:
Once upon a time, I was a Google Reader user. When it shuttered, I spent some time trying out different alternatives.
What I finally settled on was a self-hosted reader that I could access from a few different devices. But my current setup only updates my feeds four times a day, which means it only makes sense to check for new items when I know it's been updated vs. the perpetual temptation of checking Google Reader for new articles. Which changes how I approach RSS in general.
James Hawley
boosted
@lrhodes Large-scale engagement isn't an _exchange_ so much as a set of mutual public performances.
The relationship of audience-to-performer is *parasocial*, as Tom Scott (among others) notes (https://invidio.us/watch?v=leX541Dr2rU). The _audience_ feels a close bond, the _pereformer_ has no reciprocal sense.
Topology: point, pair, chain, ring, tree, star, mesh, multi-hub? At large scale, multi-hub seems to emerge, at smaller scales, others can exist. These invoke different behaviours.
James Hawley
boosted
Hello world. Bio says the important things. I'm a PhD student using biophysical techniques and computational simulation of molecular systems to understand various aspects of protein behavior and regulation. I always choose FOSS tools where possible in my work. Now lets talk *nix and science!
#introductions
#linux
#foss
#science
I'm seeing a lot of creativity from people who are diving into #rustlang, right now. Crates.io is getting a workout
New tool for making `tree` a bit more interactive
https://dystroy.org/broot/documentation/installation/
Looks pretty good, I'm giving it a spin
James Hawley
boosted
James Hawley
boosted
Hi there, I'm a science enthusiast with occupational interest in medicine, human genetics and data science. Personal interests include FOSS, science communication, open science, photography and astronomy.
I'm looking forward to inspiring discussions here on mastodon.
Ah, a new year at work, a new me to catch up on all the projects past me decided to procrastinate on
James Hawley
boosted
Does a given bacterial gene live on a plasmid or the chromosome? What other genes live on the same plasmid?
Hear how Sergio Arredondo and Anita Schürch from the University Medical Center Utrecht tackle these questions on the last bioinformatics chat of 2019:
James Hawley
boosted
Only learnt today that org-mode has checkboxes, after 4 years of using TODO, DOING, DONE for everything!
https://orgmode.org/manual/Checkboxes.html
I think most beginners overuse the nested headings in org-mode even when simple checklists are more appropriate.
I love learning new things about old tools.
* R1C1 mode for "big boy" cell references
* Ctrl+D for filling cells
* Ctrl+` for toggling between "view values" and "view formulae"
* F4 (Cmd+T on macOS) for switching between 4 different types of cell references
* "Goal seek" for changing cells to match a desired value
James Hawley
boosted
Pretty cool: Signal is now using BlurHash (https://blurha.sh), the image placeholder algorithm I developed, which is also used by Mastodon, and of course Toot!:
James Hawley
boosted
Holy shit. LaTeX goals, people:
https://castel.dev/post/lecture-notes-1/
James Hawley
boosted
#introductions I'm an immunology PhD student trying to "wrap up" / get my main project published. I've done a lot of cellular-based signaling assays in primary mouse cells, but going forward I'm very interested in functional genomics and single cell transcriptions in an immunological setting - particularly when it intersects with immunotherapies.
With my committee meeting complete, I finally had time to put the finishing touches on a quick blog post I started a while ago!
https://jrhawley.github.io/2019/12/19/well-defined
I briefly talk about the confusion around definitions in biology, and why they can cause some problems for comparing studies. Mostly, I wanted to talk about what makes a good definition and important things to think about before boldly coming up with a new one
James Hawley
boosted
@plantarum #Markdown and #Pandoc are your friends... 😉
This is a really well-written paper on #HiC data inter-sample normalization.
https://doi.org/10.1186/s12859-018-2288-x
It's not flashy, it doesn't overstate claims, it just clearly and concisely says the problems it addresses, raises the extent of its solution, gives a clear description of how it works, and performs good comparisons to demonstrate its utility.
Great work all around, we've been needing papers like this for HiC data analysis
What I want is for researchers to acknowledge that the state single-cell sequencing data analysis is still in its infancy, and many of the papers being published right now will likely not be relevant in a short period of time because they're over-stated and over-interpreted their results without bearing in mind the limitations of the measurement technique
Don't get me wrong, the technology and statistical methods behind working with this data are super cool and show lots of promise. But researchers using these hot new sequencing methods in their research can't make the bold claims they're currently making. The methodology just isn't there yet, despite people wanting it to be
My current take on the current state of #single-cell #genomics:
1. We can measure DNA/RNA from thousands of cells and identify the unique cells they come from
2. Our measurement data is noisy and sparse
3. To clean our data for analysis, we aggregate and impute information for each cell based on other cells
4. We now have semi-bulk sequencing data instead of single-cell data because we can't use the single-cell data
5. We use this semi-bulk data to identify cell types that we already knew about
- Website
- https://jrhawley.github.io
- GitHub
- https://github.com/jrhawley
- Personal Mastodon
- https://toot.cafe/@jrhawley
PhD Candidate at the University of Toronto, in the Department of Medical Biophysics.
I use my math and physics background to study computational biology in epigenetics and cancer in laboratory of Dr. Mathieu Lupien.
This is my scientific account. Views are my own.
Joined Feb 2019
