Dewei Hu

Bioinformatics | Deep learning | Ph.D. student at the University of Copenhagen

[Dewei Hu @DeweiHuX]

Looking forward to this amazing conference happening in Copenhagen this year!

[Mikaela Koutrouli @MKoutrouli]

scverse conference 2026 is coming to Copenhagen.
There’s something surreal about watching a project grow from late nights in a city you love into something that fills conference rooms across the world. And now it’s coming back to where it started for me, at least.See you there 🇩🇰

[Mikaela Koutrouli @MKoutrouli]

scverse conference 2026 is coming to Copenhagen.
There’s something surreal about watching a project grow from late nights in a city you love into something that fills conference rooms across the world. And now it’s coming back to where it started for me, at least.See you there 🇩🇰

[Dewei Hu @DeweiHuX]

3. Many thanks to my supervisors, Lars Juhl Jensen and Simon Rasmussen, and all amazing collaborators at the ZS Associates team, Anna-Lisa Schaap-Johansen, Julia Villarroel, Clara Ekebjærg, Daniel Hvidberg Hansen, and Rasmus Wernersson, for the great collaboration!

[Dewei Hu @DeweiHuX]

🎉 Pleased to share our new paper, "Molecular Maps of Diseases from Omics Data and Network Embeddings," now In Press in npj Systems Biology and Applications!

📖 Link to our paper: https://t.co/8Gyhi3Pblq

[Dewei Hu @DeweiHuX]

2. The model maps shared pathological pathways, captures tissue-specific modules, and expands predictions beyond standard assay coverage.

[bioRxiv Bioinfo @biorxiv_bioinfo]

Molecular maps of diseases from omics data and network embeddings https://t.co/8LiHgGMUOj #biorxiv_bioinfo

[Simon Rasmussen @simonrasmu]

2/
The #CCR5delta32 variant is a small deletion—32 DNA bases missing from the #CCR5 gene.
This “breaks” the CCR5 receptor, which is normally used by #HIV to infect human cells.
People who inherit this variant from both parents are highly resistant to HIV.

[Simon Rasmussen @simonrasmu]

1/
Out now in Cell: Our new study uncovers the ancient origins of a genetic mutation that protects against HIV — and rewrites the story of its surprising high frequency in Europe.
Link: https://t.co/MPNiP9cZCe

[Dewei Hu @DeweiHuX]

Glad to share our preprint: SPACE: STRING Proteins As Complementary Embeddings: https://t.co/Q41BXVvTfT

Great thanks to all the contributors: Damian Szklarczyk, Christian von Mering, and @larsjuhljensen
#bioinformatics #proteinEmbeddings #STRINGdb #computationalBiology

[Biology+AI Daily @BiologyAIDaily]

SPACE: STRING Proteins as Complementary Embeddings

• Presenting SPACE: a collection of embeddings that aligns protein network and sequence data across 1,322 eukaryotic species in the STRING database.

• Key innovation: SPACE uses the FedCoder framework to align species-specific protein network embeddings, creating a unified, cross-species embedding space while preserving species-specific network integrity.

• Complementary to sequence-based embeddings, SPACE integrates protein-protein interaction networks and orthology relations, providing a holistic view of protein functions and interactions.

• Demonstrated performance: SPACE excels in subcellular localization and protein function prediction tasks, consistently outperforming individual sequence or network embeddings in precision and recall metrics.

• Applications include cross-species predictions for subcellular localization, functional annotations, and translational research, offering a scalable, efficient tool for bioinformatics and computational biology.

• Precomputed embeddings and tools are made publicly available in STRING v12.0, providing an accessible resource for the scientific community.

@larsjuhljensen @DeweiHuX
💻Code: https://t.co/31il2ue7it
📜Paper: https://t.co/aGExOG5yos

#ProteinEmbeddings #Bioinformatics #MachineLearning #STRINGDatabase #ComputationalBiology

[Svetlana Kutuzova @sveta_kutuzova]

(1/4) Our new preprint: a metagenome binner TaxVAMB that uses taxonomic labels to create better bins. In a nutshell, TaxVAMB is great for small number of samples, incomplete genomes and non-bacterial entities @AiCentreDK @Metabolcenter https://t.co/xQflAo4ws7

[Simon Rasmussen @simonrasmu]

Join my group in beautiful Copenhagen @Metabolcenter. A 3-year international postdoc position is available in my group.

Topic is developing multi-omics foundation models

https://t.co/dpmBs9Hjb4

#jobalert #postdoc #bioinformatics #deeplearning #multimodal #integration

[Dewei Hu @DeweiHuX]

@BoWang87 Reviewers & editor (using ChatGPT): Read this paper, and give the comments.