Dee Velazquez
banner
deevelazquez.bsky.social
Dee Velazquez
@deevelazquez.bsky.social
Research assistant and software developer at JEFWorks Lab, @JHUBME. Working in #machinelearning, #computervision, and #bioinformatics. Alum @johnshopkins.
🎉 This is my first paper, and I couldn’t be more happy & excited!

A huge thank you to @jef.works for incredible guidance & mentorship, and to our reviewers for their thoughtful feedback. This has been an amazing journey! 🙌

7/7

#RStats #DataViz #SpatialTranscriptomics #OpenScience #LatinxInTech
February 25, 2025 at 3:55 PM
Try it out! 🥹

📦 CRAN: cran.r-project.org/web/packages...
💻 GitHub: github.com/JEFworks-Lab...
📚 Docs & Tutorials: jef.works/scatterbar/

6/7
February 25, 2025 at 3:52 PM
🐭 Case Study: Mouse Brain Visium ST Data

We visualized deconvolved cell-type proportions in an adult mouse brain using scatterpie vs. scatterbar. The result? Scatterbar made proportional differences easier to interpret than scatter pie plots! 🧠🎉

5/7
February 25, 2025 at 3:50 PM
🛠️ We need a better approach!

Enter scatterbar 📊✨

Instead of pie charts, scatterbar uses stacked bar plots:
✅ More precise comparisons using bar lengths
✅ Better use of space with aligned bars
✅ Enhanced visual saliency for detecting subtle changes

4/7
February 25, 2025 at 3:49 PM
📊 Scatter pie plots are common but flawed. Why?

Pie charts encode data using angles, which makes comparisons harder than using bar lengths—especially for subtle differences. Studies show bars are more readable and accurate than pie slices, so we need a better approach!

3/7
February 25, 2025 at 3:49 PM
🔬 Why does this matter?

#SpatialTranscriptomics (ST) enables high-throughput gene expression profiling at multi-cellular pixel resolution. Accurately visualizing cell-type proportions & spatial changes is crucial for understanding tissue organization & disease mechanisms.

2/7
February 25, 2025 at 3:48 PM