Kalin Nonchev
@nonchevk
Followers
83
Following
112
Media
19
Statuses
64
PhD at @ETH Zurich, machine learning and biomedical data
Zurich, Switzerland
Joined March 2022
How can we predict spatial transcriptomics from histology images to enable simple, affordable, and reliable analysis of spatially resolved gene expression in routine clinical use? 🤔 Introducing DeepSpot – a deep learning model designed to tackle this! 🧵👇
1
0
6
@gxr ✉️Full job description and how to apply: https://t.co/CXzfQpsq9X Application ❗️Applications will be considered only if submitted through the specified process, and incomplete applications will not be considered.
0
0
0
Join us for an exciting internship where cutting-edge machine learning research meets real-world biomedical data!
1
0
0
Internship Opportunity: Multimodal AI Research Scientist at the Biomedical Informatics Group at ETH Zurich 🚀 Interested in working at the intersection of computational pathology, spatial transcriptomics, LLM representation learning, and tissue generation?
1
2
2
We just released Hi-C predictions for 157 ENCODE ATAC-seq cell lines and primary cells, made using our model UniversalEPI 🎉 From ATAC-seq to genome architecture, no Hi-C experiment required! Super excited to see these tracks now live on the @GenomeBrowser! Start exploring👇
We are pleased to announce our latest public hub, UniversalEPI ENCODE for hg38. It shows Hi-C interaction predictions based on ENCODE ATAC-seq data, generated by UniversalEPI: https://t.co/WW5nviE6j3 Thanks to the Boeva Lab at ETH Zurich for creating this hub.
1
7
27
Many thanks for the invitation and for the opportunity to experience and learn more about Chinese culture.
0
0
0
It is fascinating to learn about the real-world challenges they face in multi-omics and I had some engaging conversations about how we can help address them using our biomedical machine learning expertise.
1
0
0
Winning first place at MICOS EU 2024 with our machine learning model for representation learning in spatial transcriptomics (AESTETIK: https://t.co/P5jePdZWDx) led to an incredible opportunity to visit BGI Genomics’ headquarters in Shenzhen, China.
medrxiv.org
Spatial transcriptomics enables in-depth molecular characterization of samples on a morphology and RNA level while preserving spatial location. Integrating the resulting multi-modal data is an...
BGI Group concluded the DUT-BGI International Summer School 2025, uniting 38 global students for hands-on genomics and innovation workshops, empowering future leaders and fostering international collaboration in life sciences and health. https://t.co/t9Fit2x171
1
0
0
Just presented our new multimodal histopathology method "SpotWhisperer" at ICML, one of the largest AI conference. SpotWhisperer enables spatially resolved annotation of histopathology images using natural language by "transferring" annotations from transcriptomic data.
🔬 Toward histopathology 2.0: spatial transcriptomes inferred from routine diagnostic H&E images + a chat interface for cell-resolution histopathology through English language. (1/6)
0
2
6
🔬 Toward histopathology 2.0: spatial transcriptomes inferred from routine diagnostic H&E images + a chat interface for cell-resolution histopathology through English language. (1/6)
1
11
34
With DeepSpot, Kalin Nonchev bridges high-performance ML with clinical relevance, making spatial biology more accessible for research and potentially patient care. One of the most impactful submissions in the Autoimmune Disease ML Challenge.
1
1
2
Learn more about DeepSpot, developed at ETH Zürich: https://t.co/gJJA3HjL5k Code: https://t.co/dXf6968kyV Virtual spatial transcriptomics:
medrxiv.org
Spatial transcriptomics technology remains resource-intensive and unlikely to be routinely adopted for patient care soon. This hinders the development of novel precision medicine solutions and, more...
0
0
0
DeepSpot addresses this by using transfer learning to integrate virtual spatial transcriptomics with Inflammatory Bowel Disease single-cell atlas, enabling accurate prediction of previously unobserved biomarkers.
1
0
0
A major limitation of single-cell spatial transcriptomics is the limited number of detected genes.
1
0
0
Next important milestone in validating our work to broaden access to spatial transcriptomics and support progress in clinical stratification and targeted therapies. Many thanks to @broadinstitute and @crunchDAO for hosting these competitive ML challenges and raising awareness.
Presenting Kalin Nonchev’s (@nonchevk) DeepSpot submission for the Autoimmune Disease ML Challenge II with @Schmidt_Center It improved gene correlation across datasets from patients with metastatic melanoma, kidney, lung & colon cancers beating the standard benchmark 🧵
1
0
0
0
0
0
Explore the dataset: https://t.co/MWzvPTtP8p Manuscript: https://t.co/5BOTYo9hzU GitHub:
huggingface.co
1
0
0
💡 This resource unlocks exciting opportunities for developing new multi-modal deep learning methods, benchmarking existing ones, and accelerating biological discoveries in cancer research using digital spatial transcriptomics.
1
0
0