Update research and publications

pages/publications.md

@@ -6,14 +6,16 @@ permalink: /publications
 
 For an up to date list of all publications (including preprints), please see [Google scholar](https://scholar.google.com/citations?user=H4jO_DQAAAAJ&hl=en).
 
-[2024] (#2024) [2023](#2023) [2022](#2022) [2021](#2021) [2020](#2020) [2019](#2019) [2018](#2018) [2017](#2017) [2016](#2016)
+[2024](#2024) [2023](#2023) [2022](#2022) [2021](#2021) [2020](#2020) [2019](#2019) [2018](#2018) [2017](#2017) [2016](#2016)
 
 ## <a id="2024"></a>2024
 
 1. Wei E. Gordon, Seungbyn Baek, Hai P. Nguyen, Yien-Ming Kuo, Rachael Bradley, Sarah L. Fong, Nayeon Kim, Alex Galazyuk, Insuk Lee, Melissa R. Ingala, Nancy B. Simmons, Tony Schountz, Lisa Noelle Cooper, Ilias Georgakopoulos-Soares, **Martin Hemberg**, Nadav Ahituv, [* Integrative single-cell characterization of a frugivorous and an insectivorous bat kidney and pancreas*](https://www.nature.com/articles/s41467-023-44186-y), **Nature Communications**, 15 (12), January 2024.
 
 2. Yufei Wang, Jae-Won Cho, Gabriella Kastrunes, Alicia Buck, Cecile Razimbaud, Aedin C Culhane, Jiusong Sun, David A Braun, Toni K Choueiri, Catherine J Wu, Kristen Jones, Quang-De Nguyen, Zhu Zhu, Kevin Wei, Quan Zhu, Sabina Signoretti, Gordon J Freeman, **Martin Hemberg**, Wayne A Marasco, [*Immune Restoring CAR-T Cells Display Antitumor Activity and Reverse Immunosuppressive TME in a Humanized ccRCC Mouse Model*](https://www.cell.com/iscience/pdf/S2589-0042(24)00100-7.pdf), **iScience**, 2024.
 
+3. Jooyeon Suh, Hyeonkyeong Kim, Jiyun Min, Hyun Ju Yeon, **Martin Hemberg**, Luca Scimeca, Ming-Ru Wu, Hyun Guy Kang, Yi-Jun Kim, Jin-Hong Kim, [*Decoupling NAD + metabolic dependency in chondrosarcoma by targeting the SIRT1-HIF-2a axis*](https://www.cell.com/cell-reports-medicine/pdf/S2666-3791(23)00559-1.pdf), **Cell Reports Medicine**, Volume 5, 101342, 2024.
+
 ## <a id="2023"></a>2023
 
 1. Ioannis Mouratidis, Candace S Y Chan, Nikol Chantzi, Georgios Christos Tsiatsianis, **Martin Hemberg**, Nadav Ahituv, Ilias Georgakopoulos-Soares, [*Quasi-prime peptides: identification of the shortest peptide sequences unique to a species*](https://academic.oup.com/nargab/article/5/2/lqad039/7138487), **NAR Genomics and Bioinformatics**, Volume 5, Issue 2, June 2023.
@@ -22,7 +24,6 @@ For an up to date list of all publications (including preprints), please see [Go
 
 3. Jacob Hepkema, Nicholas Keone Lee, Benjamin J Stewart, Siwat Ruangroengkulrith, Varodom Charoensawan, Menna R Clatworthy, **Martin Hemberg**, *[Predicting the impact of sequence motifs on gene regulation using single-cell data*](https://genomebiology.biomedcentral.com/articles/10.1186/s13059-023-03021-9), **Genome Biology**, Volume 24, Issue 1, p 1-24, 2023.
 
-4. Jooyeon Suh, Hyeonkyeong Kim, Jiyun Min, Hyun Ju Yeon, **Martin Hemberg**, Luca Scimeca, Ming-Ru Wu, Hyun Guy Kang, Yi-Jun Kim, Jin-Hong Kim, [*Decoupling NAD + metabolic dependency in chondrosarcoma by targeting the SIRT1-HIF-2a axis*](https://www.cell.com/cell-reports-medicine/pdf/S2666-3791(23)00559-1.pdf), **Cell Reports Medicine**, Volume 5, 101342, 2024.
 
 ## <a id="2022"></a>2022
 

pages/research.md

@@ -13,44 +13,44 @@ The cell is the fundamental unit of biology, and long term changes involve a coo
 
 We are interested in developing novel computational methods for analyzing high throughput sequencing data. In particular, there has been a focus on single-cell RNAseq, but we are also interested in other types of sequencing data.
 
-### Gene network inference using scRNAseq data
+### Quantitative metrics for analyzing cellular architecture
 
-Since the introduction of microarrays, scientists have tried to use genome wide transcriptome data to infer gene-gene interactions. With scRNAseq data it becomes possible to identify groups of cells of the same type which should in principle make this task easier. Nevertheless, the data remains noisy and in this project in collaboration with the [Lee](https://netbiolab.org/w/Welcome_to_Network_Biology_Laboratory) lab at Yonsei University in South Korea we are using a reference guided approach to leverage the large number of interactions that have been documented in the literature.
+![Pattern finding in spatial transcriptomics](/assets/img/spatialscfind_example.jpg)
 
-### Fast unsupervised clustering
+The latest generation of spatial transcriptomics methods make it possible to profile cells in their tissue context. The high-resolution spatial information allows us to investigate how cells are organized in the tissue. We are using ideas from spatial statistics to develop new approaches for understanding the organizational principles of cellular architecture. These methods make it possible to quantify differences in tissue structure, and will be helpful for understanding the impact of various disease conditions.
 
-![SC3s flow chart](/assets/img/sc3s.png)
+### Batch correction and data integration
 
-In 2017, we published the single cell consensus clustering [method](https://github.com/hemberg-lab/SC3) (SC3) which is a robust and accurate method for unsupervised clustering of scRNAseq data. The main shortcoming of the algorithm, however, is its poor scalability to samples with more than a few thousand cells. The new [SC3 with speed](https://github.com/hemberg-lab/sc3s) (SC3s) improves several bottlenecks in the algorithm which allows it to scale linearly in time and memory with the number of cells.
+In collaboration with the [Korsunsky](https://www.korsunskylab.org/) and [Raychaudhuri](https://immunogenomics.hms.harvard.edu/) labs we are developing the next version of the popular [Harmony](https://github.com/immunogenomics/harmony) package for merging single cell datasets to eliminate technical artefacts. The new version is orders of magnitudes faster and more memory efficient, allowing for the integration of tens of millions of cells from tens of thousands of batches. In addition, the upgraded method features automatic tuning of hyper parameters to avoid overcorrection.
 
-### Coordinated splicing events in scRNAseq data
+### Inferring cell-cell communication from high-resolution spatial transcriptomics data
 
-Protocols that allow for full-length coverage of transcripts make it possible to investigate the utilization of different splice junctions at the single-cell level. Even though isoform quantification is a challenging problem given the protocols used today, it is possible to accurately assess the usage of splice junctions. By building a data structure that allows for fast queries of splice junction usage across all cells profiled in a scRNA-seq experiment, we can systematically search for cell-type specific splicing patterns, e.g. flip-flop exons and other co-regulated splicing events. Moreover, it will allow us to identify cell-type specific splicing events, akin to marker genes.
+We are developing [Scotia](https://github.com/Caochris/SCOTIA), a method for inferring cell-cell interactions via ligand-receptor communication. Scotia can leverage high-resolution spatial transcriptomics data to evaluate both the distance between cells as well as their similarity in terms of gene expression. As such, it is possible to exclude interactions that are implausible due to cells being located far apart. Predictions from Scotia for pancreatic adenocarcinoma have been experimentally validated, suggesting that it can identify relevant interactions. 
 
-### Quantitative models of regulatory motifs at promoters and enhancers
+### Early detection of cancer using cell-free RNA
 
-
-![Example scover analysis](/assets/img/scover.png)
-
-Building on our recent work using convolutional neural networks for identification and quantification of regulatory motifs, we will expand the model to study the impact of distal enhancers as well. This will allow us not only to discover regulatory elements that are preferentially found at either promoters or enhancers, but also to quantify their impact on gene expression. Furthermore, we also seek to incorporate non-coding genetic variants.
+Liquid biopsies have emerged as a cost-effective and minimally invasive method for detecting cancer at an early stage. Through a simple blood draw, it is possible to detect molecules that have been shed from a tumor. To date, most work has focused on cell free DNA, but it is also possible to detect RNA molecules from the tumor. We are developing computational methods for detecting such signals from next generation sequencing data, and an important advantage of RNA over DNA is that it can also provide useful information about the malignant transcriptome.
 
 ## Collaborative Projects
 
-We have worked with several experimental group in a wide range of areas including neuroscience, immunology and cancer.
+We work with several experimental groups, and most of our collaborations are in the areas of neuroscience, immunology and cancer.
 
-### Perianal fistulizing disease
+### Identification of drivers of treatment resistance in pancreatic cancer
 
-We are working with the [Salas lab](https://www.clinicbarcelona.org/en/idibaps/research-areas/liver-digestive-system-and-metabolism/inflammatory-bowel-disease) at IDIBAPS in Barcelona to profile patients with this subtype of inflammatory bowel disease. It is our hope that single-cell profiling will make it possible to identify biomarkers and therapeutic targets.
+![PDAC_interactions](/assets/img/scotia.jpg)
 
-### Characterization of sequences absent from the genome
+In collaboration with the [Hwang](https://www.whwanglab.org/) lab we are working on multiple projects related to pancreatic cancer, with a focus on understanding what are the drivers of therapeutic resistance. This is a common problem with this malignancy, and one of the main reasons behind the high mortality rates. We are analyzing high-resolution, high-plex spatial transcriptomics and spatial proteomics data to characterize the microenvironment, to understand how it varies across different malignant subtypes. 
 
-![Neomers heatmap](/assets/img/neomers_heatmap.png)
+### Determinants of influenza vaccine protection
 
-In collaboration with the [Ahituv lab](https://pharm.ucsf.edu/ahituv) we are investigating [nullomers](https://en.wikipedia.org/wiki/Nullomers), short k-mers that are *absent* from the genome. In our first publication we have characterized both nullomers and nullpeptides across 30 species as well as across human populations, and in a subsequent [preprint](https://www.medrxiv.org/content/10.1101/2021.08.15.21261805v1) we demonstrate how these sequences can be used as cancer biomarkers. Importantly, this includes liquid biopsies which implies that no tumor DNA is required.
+We are working with the [Marasco](https://marascolab.dana-farber.org/) lab to investigate what are the molecular mechanisms underlying a strong immune protection following vaccination against the seasonal flu. This is particularly important for the elderly population where protection rates are the lower and mortality is the highest. By profiling PBMCs via single cell RNAseq following vaccinations, we aim to identify cell types and genes that are associated with a favorable response.
 
+### Perianal fistulizing disease
+
+We are working with the [Salas lab](https://www.clinicbarcelona.org/en/idibaps/research-areas/liver-digestive-system-and-metabolism/inflammatory-bowel-disease) at IDIBAPS in Barcelona to profile patients with this subtype of inflammatory bowel disease. It is our hope that single-cell profiling will make it possible to identify biomarkers and therapeutic targets.
 
 ### CAR T-cells for solid tumors
 
 ![CART UMAP](/assets/img/cart_umap.png)
 
-Together with the [Marasco lab](https://marascolab.dana-farber.org/) at the Dana-Farber Institute we are trying to understand the mechanisms involved in a novel design which is aimed at overcoming the suppresive tumor microenvironment. We are using scRNA-seq to profile the immune microenvironment.
+Together with the [Marasco lab](https://marascolab.dana-farber.org/) at the Dana-Farber Cancer Institute we are trying to understand the mechanisms involved in a novel design which is aimed at overcoming the suppresive tumor microenvironment. By profiling tumors using single cell RNAseq and spatial transcriptomics we can identify the cellular and molecular features that distinguish poor and good responders.