Lissa Cruz-Saavedra

Postdoctoral researcher. McGill university. Trypanosoma cruzi, my passion. Travel, my life.

[Nature Biotechnology @NatureBiotech]

Survey data document US visa types and burdens of time, cost and mental health among international postdocs https://t.co/JMrhrxTHlB
https://t.co/Qv9BtHqn0A

[parasitesrule @parasitesrule]

🔬 Featured Image
Trypanosoma cruzi amastigotes (red) infecting a cardiomyocyte (cyan) in a mouse heart, with CD4 (green) and CD8 (magenta) T cells revealing the immune response.

📸 Fernando Sanchez-Valdez, @CTEGD, University of Georgia
https://t.co/lrjDRvK7TX

[Hasan Toor @hasantoxr]

🚨BREAKING: Google just dropped another hit!

It's called PaperBanana and it generates publication-ready academic illustrations from just your methodology text.

No Figma. No manual design. No illustration skills needed.

Here's how it works:

A team of AI agents runs behind the scenes
→ One finds good diagram examples
→ One plans the structure
→ One styles the layout
→ One generates the image
→ One critiques and improves it

Here's the wildest part:

Random reference examples work nearly as well as perfectly matched ones. What matters is showing the model what good diagrams look like, not finding the topically perfect reference.

In blind evaluations, humans preferred PaperBanana outputs 75% of the time.

This is the recursion we've been waiting for AI systems that can fully document themselves visually.

Waitlist’s open, Link in the first comment.

[Ronald van Loon @Ronald_vanLoon]

#AI Tools for Literature Review
by @Faheem_uh

#ArtificialIntelligence #Innovation #EmergingTech #TechForGood

cc: @marcusborba @HaroldSinnott @karpathy

[Biology+AI Daily @BiologyAIDaily]

Democratizing Protein Language Model Training, Sharing and Collaboration @NatureBiotech

1. A groundbreaking initiative, ColabSaprot and SaprotHub, has been introduced to democratize the training and sharing of protein language models (PLMs), making advanced AI techniques accessible to researchers without deep machine learning expertise.

2. The platform, built on Google Colab, allows researchers to train their own task-specific PLMs through an intuitive interface, supporting a wide range of prediction tasks from protein function prediction to protein sequence design.

3. A key innovation is the introduction of a novel structure-aware alphabet (SAA) that encodes both amino acid type and local geometry, addressing scalability and overfitting challenges in training large-scale PLMs.

4. The foundation model, Saprot, demonstrates superior performance across diverse protein prediction tasks compared to existing models like ESM-2 and ProtBert, especially in tasks involving structural information.

5. ColabSaprot integrates lightweight adapter networks, enabling efficient fine-tuning and sharing of models within the research community, reducing storage and communication burdens.

6. SaprotHub serves as a community repository for storing, sharing, and collaboratively developing fine-tuned models, fostering a collaborative ecosystem where researchers can build on each other’s work.

7. Experimental validations in wet labs have shown promising results, with several predicted protein variants demonstrating enhanced functionality, highlighting the practical utility of the platform.

8. A user study involving biologists without ML backgrounds demonstrated that ColabSaprot and SaprotHub enable them to train and use state-of-the-art PLMs with performance comparable to AI experts.

💻Code: https://t.co/sjxZvUVmxQ
📜Paper: https://t.co/Mzfx8qVADg
#ProteinLanguageModel #AIinBiology #OpenScience #MachineLearning #ProteinEngineering #CollaborativeResearch

[Biology+AI Daily @BiologyAIDaily]

RFdiffusion2 is open sourced!
https://t.co/QafvOWCcdR

[Prof Lennart Nacke, PhD @acagamic]

Every research objective needs to answer five questions:

• Why does this matter?
(Significance)

• How will you do it?
(Methodology)

• When will it be done?
(Feasibility)

• Who benefits?
(Relevance)

• What data will you use?
(Evidence)

Miss any one, and your project is built on shaky ground.

[Lissa Cruz-Saavedra @LissaCruzi]

Feeling accomplished after completing a workshop focused on single-cell RNA-seq 🧬🧠
So much new knowledge, powerful tools, and fresh ideas to take forward.
Excited to explore all the possibilities ahead in single-cell biology in #Trypanosomacruzi!

[Alex Zhavoronkov, PhD (aka Aleksandrs Zavoronkovs) @biogerontology]

My dear friends, I have a great story to share. After a night of hard work, one of my agents working on real-time monitoring of publications related to targets of interest informed me of a high-profile publication on TNIK. And when I clicked on the link, I almost fell out of my chair - it was a detailed commentary on rentosertib explaining the program in detail by one of the heroes of the generative AI revolution in biomedicine and someone who (and who's team) I deeply respect, @marinkazitnik ! 🙏🙏
I usually celebrate with a hackathon on novel target discovery, and tomorrow we will be spending half a day with our AI platforms to come up with a perfect target for the next big program 😊
Please check out the commentary and read the two papers documenting the history of discovery and development in the comments.

hashtag#GreatDay hashtag#ResearchPaper hashtag#AI

[César de la Fuente @delafuentelab]

Ever wondered how to spot new antibiotics hiding in biological data? Our latest paper in @NatureProtocols @NaturePortfolio offers a practical, end-to-end ML playbook for mining genomes & proteomes—AI vs AMR, one dataset at a time.

Link: https://t.co/pXIemucmFS

[CIMBIUR @CIMBIUR]

Lissa Cruz, bacterióloga apasionada por la parasitología 🧫🔬, formada en CIMBIUR 🏛️. Con maestría y doctorado en Ciencias Biomédicas 🎓, hoy investiga Trypanosoma cruzi 🦠 en McGill 🇨🇦. Su misión: aportar soluciones y visibilizar la enfermedad de Chagas endémico de Colombia 🇨🇴

[Ming "Tommy" Tang @tangming2005]

1/ Bioinformatics takes years to master.
Not because it’s hard.
But because so much of what matters… no one writes down.
Let me explain

[AJ Thurston, PhD @AJThurston]

Me coming back to R after bravely venturing into Python.

[VEuPathDB @veupathdb]

Keep VEuPathDB open-access to all by subscribing today! https://t.co/BCvvTkwvEb

[Lissa Cruz-Saavedra @LissaCruzi]

Excited to share our latest preprint: "Stochastic variation in surface protein expression diversifies Trypanosoma cruziinfection" 🧬
We combined a new chromosome-level genome, TMT proteomics, MGF abundance profiling & a broad antibody library.
🧪 https://t.co/0Qx2bxwUPq
#Chagas

[parasitesrule @parasitesrule]

Do you work on parasites and reside in an LMIC? Consider applying for a 2025 MPM LMIC award. The deadline is March 3rd, 2025—details on the website.
https://t.co/fsZx5aOho8

[Luisa Figueiredo @luisamfigueired]

Have you finished your #PhD and are interested in exploring how #parasites cross the vasculature to colonize tissues? Do you fancy #surfing?

Join us in #Lisbon at @GIMM_Institute , where science meets quality of life.

@parasitesrule @CaixaResearch @JobsScience @EMBO

[URosario @URosario]

El Colegio Elector ha elegido a Ana Isabel Gómez Córdoba como nueva rectora de la Universidad del Rosario para el periodo 2024-2026, por primera vez una mujer asumirá el liderazgo de la Universidad a partir del 24 de octubre. "Es una inmensa responsabilidad, más aún cuando por primera vez en la historia de nuestro Claustro le corresponde a una mujer”

[VEuPathDB @veupathdb]

Welcome back! VEuPathDB resources are back online today! We will continue to provide bioinformatics services under a new organizational structure. Read more here: https://t.co/YNX4KBuIxV

[Reviewer 2 @GrumpyReviewer2]

#AcademicChatter #AcademicX