Delayed, but with bug fixes...🤷🏾 Life Update (v3.0): - Since January 2022, I am a Ph.D. candidate Helmholtz AI & Saur Lab, TU Munich. - I research multi-modal data to unravel certain mysteries of cancer. - I moved to Germany!

We are very excited to announce scverse (scverse.org), a new consortium around the core Python packages for single-cell omics data analysis. scverse is a cross-lab effort to ensure the longevity and interoperability of the single-cell analysis ecosystem in Python.

Excited to share our latest work: Combining #graphs and #transformers for robust weakly #supervised learning for whole slide image analysis. Published IEEE-TMI (doi.org/10.1109/TMI.20…) with YZ, Kolachalama laboratory, Jennifer Beane BU CompBioMed, Boston U Chobanian & Avedisian School of Medicine, BU AIR, Hariri Institute for Computing, Boston University

Good research is an outcome of believing you have freedom at all hierarchical levels of a "thinking conscious & subconscious mind".

Looking forward to connecting with a few smart researchers and understand the interesting projects using AI at #HelmholtzAIcon2022.

Another glorious preprint from the Johns Hopkins Kimmel Cancer Center Convergence team (Dr. Luciane T Kagohara 🇧🇷🧬👩🏻‍🔬👩🏻‍💻 et al): Spatial transcriptomics of FFPE pancreatic intraepithelial neoplasias reveals cellular and molecular alterations of progression to #PancreaticCancer biorxiv.org/content/10.110…

How to develop a productive routine? When transiting from structured learning environments (e.g., courses) to unstructured ones (e.g., research), developing a productive routine and managing their time well are critical for productivity. Sharing some hacks I found useful! 👇

Nice!!

As a computational researcher, I am curious to know how much of a variation an expert pathologist can observe in the cellular morphology with changes in it's genomic content? Is that morphological variation measurable? Saur Lab would you like to share your thoughts?

As PhD application deadlines approach, we are super excited to announce cs-sop.org, created by Zhaofeng Wu Alexis Ross @ZejiangS💻 cs-sop.org is a platform with statements of purpose generously shared by previous applicants to CS PhD programs 🧵(1/n)

Some advice I give my graduate students about how to keep on top of the literature in their field. A 🧵: 1/

Now that we have a pretty cool documentation for sc data analysis best practices, it's time we step up the game and develop spatial data analysis tools...

😍😍 our paper on scale invariance in pre trained CNNs in the top 5 of MAKE!

Can't agree more! Also, learning about biology "empirically" can be difficult + not reflect the real world scenario. We should focus more on interpretability and robustness!

Good luck to all the PhD students this week, where all their family try to understand what they do and ask when they will be done.

Glad to see our review on best practices for single-cell analysis across modalities out Nature Reviews Genetics! In a big team effort led by Lukas Heumos & Anna Schaar, we recommend workflows based on benchmarks. Paper at rdcu.be/c8RU5 & extension at sc-best-practices.org.

Our paper presenting scverse - our multi-institution open-source software project to address storage and analysis needs of single-cell data - is now online in Nature Biotechnology! nature.com/articles/s4158… & rdcu.be/c9yNV.

Is it necessary to publish every single project as a python package? I realise the importance of publishing projects as executable package, but I notice good projects breaking because of a cascade failure of out of date dependencies from other "small" packages.....Any thoughts?

"Those are some interesting facts! In my experience, a career in Informatics Research hasn't been as common in South Asian countries, possibly influenced by cultural factors. It's great to see things different today even if it’s a different field of science!" 🙌🤩