Cui Research Group

Are you a passionate, forward-thinking individual with a knack for computer science and an insatiable curiosity for the world of biology and medicine? Join the Cui Research Group in the exciting field of Artificial Intelligence (AI) and Biology.

About Feng Cui

headshot of Feng CuiFeng Cui, Ph.D. is an Associate Professor in the RIT Thomas H. Gosnell School of Life Sciences. He earned a Ph.D. in Bioinformatics and Computational Biology at Iowa State University and subsequently spent several years as a postdoctoral fellow at National Cancer Institute (NCI). His research is mainly focused on the nucleosome and its associated binding proteins. Dr. Cui serves as the Graduate Director of the Bioinformatics MS Program and holds the position of Faculty Senator (Alternate). He enjoys working with students at the interface of Biology and Computer Science. Dr. Cui is also an affiliated faculty member of the Golisano College for Computing and Information Sciences.

Seeking Undergraduate and Graduate Students

The Cui Research Group is seeking BS, MS and Ph.D. students who demonstrate a passion for learning and possess coding abilities. As we continue to push the boundaries of knowledge in our field, we invite you to join us and make a valuable contribution to our cutting-edge research. Do you have the following requirements and want to work in a dynamic and collaborative environment? 

  • Demonstrated coding skills.
  • Interest in gaining valuable experience.
  • Enthusiasm for learning and research.
  • Commitment to the project.

If you meet these requirements we encourage you to apply for a position in our lab

Current Projects

1. Exploring the diversity of nucleosomal DNA

The basic repeating unit of chromatin is the nucleosome core particle, which consists of a histone octamer, around which 147 bp DNA are wrapped about 1.7 turns. Structurally, the 147-bp DNA fragment can be divided to two halves by the dyad at position 74. Each half is consisted of six minor-groove bending sites (GBS) (at superhelical locations (SHL) -6.5 to -1.5) where DNA is bent into minor grooves and six major-GBS (SHL -6 to -1) where DNA is bent into major grooves.

Four types of DNA patterns have been observed in nucleosomal DNA. Type 1 is the pattern described by Travers and colleagues, in which WW dimers have peaks at minor-GBS (blue shading), and SS dimers have peaks at major-GBS (red shading). Type 2 is the pattern in which both WW dimers and SS dimers have peaks at minor-GBS. Type 3 is the pattern in which both WW dimers and SS dimers have peaks at major-GBS. Type 4 is opposite to Type 1, in which WW dimers have peaks at major-GBS, and SS dimers have peaks at minor-GBS. Here, we refer to Type 1 as the well-known ‘WW/SS pattern’ and Type 4 as the ‘anti-WW/SS pattern’.

Our work aims to understand the structural basis and functional significance of the nucleosomes with Type 4 pattern (i.e., anti-WW/SS nucleosomes). We have found that anti-WW/SS nucleosomes account for 13-31% of all nucleosomes in different species and are widespread across the genomes. They are enriched in genic regions in mammals and correlated with transcriptional levels. Currently, we are trying to understand whether they are also enriched around transcription factor binding sites and how their structures are stabilized.

diagram of nucleosomal DNA

2. Predicting nucleosome-binding proteins by machine learning

A small group of TFs known as pioneer TFs are able to bind nucleosomal DNA and play an important role in cell development. Based on the Systematic Evolution of Ligands by Exponential Enrichment (SELEX) technique, recent studies have identified several TF-nucleosome interaction modes including end binding, oriented binding, periodic binding, dyad binding, groove binding, and gyre spanning. However, there are substantial experimental challenges in measuring nucleosome binding modes for thousands of TFs in different species. In this study, we aimed to test a hypothesis that the binding modes of a TF to a nucleosome can be predicted by their amino acid sequences using machine learning methods. To achieve this aim, we have developed a machine learning model, ProtGauss, to predict the nucleosome binding modes based on an unsupervised data-driven Gaussian representation for protein sequences. Currently, we are developing novel deep learning frameworks to predict nucleosome-binding proteins in a specie-specific manner.

diagram of nucleosome-TF binding patterns

3. AI in medicine and systems biology

The biological system is a complex network of heterogeneous molecular entities such as genes, proteins, and other biomolecules linked together by their interactions. Although enormous technological advancements have been made in the past four decades, experimental determination of these interactions has been a great challenge. With the ever-increasing computational resources and novel methodologies capable of handling large datasets, we developed a suite of algorithms based on deep learning to represent biological entities and model their interactions. These methods accurately predicted the interactions between genes, between proteins, and between drugs and targets. Moreover, we developed novel deep learning frameworks to predict bladder cancer types, smooth muscle fibers, as well as biomarkers associated with distant metastasis of various carcinoma. Currently, we are developing novel AI models to predict novel sex-associated functional biomarkers of bladder cancers. Moreover, we are investigating the role of ChatGPT in omics research.

diagram of prediction of bladder muscle fibers

Research reported on this page was supported by National Institute of General Medical Sciences (NIGMS) of the National Institutes of Health under award number R15GM149587.

Latest News

Current Members

headshot of Isaac Olatunji
Isaac Olatunji 

headshot of Sanjeev Vijayakumar
Sanjeev Vijayakumar

Sridevi Kayyur Subramanya '22 (bioinformatics MS)
Data Science Analyst at Navigate BioPharma Services, Inc.

Sheethal Umesh Nagalakshmi '21 (bioinformatics MS)
Computational Biologist at Beth Israel Deaconess Medical Center

Andrew Rosato '21 (bioinformatics MS)
Associate Bioinformatician at Massachusetts General Brigham

Peng Nien Yin '19 (bioinformatics MS)
Machine Learning Engineer at Paychex

Jimmy Zhang '18 (bioinformatics MS)
Senior Software Engineer at Athena Health

Feifei Bao '16 (bioinformatics MS)
Bioinformatician at Houston Methodist

Julia Freewoman '16 (bioinformatics MS)
Adjunct Lecturer at RIT

Peter LoVerso '15 (bioinformatics MS)
Bioinformatician at Precision Diagnostics

Gregory Wright '15 (bioinformatics MS)
Ph.D. student at University of Colorado

Bader A. Alharbi '14 (bioinformatics MS)
Ph.D. student at George Mason University


Journal Paper
Olatunji, Isaac and Feng Cui. "Multimodal AI for prediction of distant metastasis in carcinoma patients." Frontiers in Bioinformatics 3. (2023): 1131021. Web.
Subramanya, Sridevi K., et al. "Deep learning for histopathological segmentation of smooth muscle in the urinary bladder." BMC Medical Informatics and Decision Making 23. (2023): 122. Web.
Rynkiewicz, Patrick, et al. "Functional binding dynamics relevant to the evolution of zoonotic spillovers in endemic and emergent Betacoronavirus strains." Journal of Biomolecular Structure and Dynamics 40. 21 (2022): 10978-10996. Web.
Kc, Kishan, et al. "Predicting Biomedical Interactions with Higher-Order Graph Convolutional Networks." IEEE/ACM Transaction on Computational Biology and Bioinformatics 19. 2 (2022): 676-687. Web.
Gupta, Aditya, Andrew J. Rosato, and Feng Cui. "Vaccine candidate designed against carcinoembryonic antigen-related cell adhesion molecules using immunoinformatics tools." Journal of Biomolecular Structure and Dynamics 39. 16 (2021): 6084–6098. Web.
Freewoman, Julia M, Rajiv Snape, and Feng Cui. "Temporal gene regulation by p53 is associated with the rotational setting of its binding sites in nucleosomes." Cell Cycle 20. 8 (2021): 792-807. Web.
Kc, Kishan, et al. "Machine learning predicts nucleosome binding modes of transcription factors." BMC Bioinformatics 22. (2021): 166. Web.
Yin, Peng-Nien, et al. "Histopathological distinction of non-invasive and invasive bladder cancers using machine learning approaches." BMC Medical Informatics and Decision Making 20. 1 (2020): 162. Web.
Zhang, Jimmy F., et al. "BioVR: a platform for virtual reality assisted biological data integration and visualization." BMC Bioinformatics 20. (2019): 78. Web.
KC, Kishan, et al. "GNE: a deep learning framework for gene network inference by aggregating biological information." BMC Systems Biology 13. (2019): 38. Web.
Wright, Gregory M and Feng Cui. "The nucleosome position-encoding WW/SS sequence pattern is depleted in mammalian genes relative to other eukaryotes." Nucleic Acids Research 47. 15 (2019): 7942–7954. Print.
F., Bao, et al. "P53 Binding Sites in Normal and Cancer Cells are Characterized by Distinct Chromatin Context." Cell Cycle 16. 21 (2017): 2073-2085. Print.
Cole, Hope A., et al. "Novel Nucleosomal Particles Containing Core Histones and Linker DNA but no Histone H1." Nucleic Acids Research 44. 2 (2016): 573-581. Print.
Ocampo, Josefina, et al. "The Proto-chromatosome: A Fundamental Subunit of Chromatin?" Nucleus 7. 4 (2016): 382-387. Print.
LoVerso, Peter R and Feng Cui. "A Computational Pipeline for Cross-Species Analysis of RNA-seq Data Using R and Bioconductor." Bioinformatics and Biology Insights 9. (2015): 165-174. Print.
LoVerso, Peter R, Christopher M Wachter, and Feng Cui. "Cross-species Transcriptomic Comparison of In Vitro and In Vivo Mammalian Neural Cells." Bioinformatics and Biology Insights 9. (2015): 153-164. Print.
Norouzi, Davood, et al. "Topological diversity of chromatin fibers: Interplay between nucleosome repeat length, DNA linking number and the level of transcription." AIMS Biophysics 2. 4 (2015): 613-629. Print.
Cui, Feng and Victor B. Zhurkin. "Rotational Positioning of Nucleosomes Facilitates Selective Binding of p53 to Response Elements Associated with Cell Cycle Arrest." Nucleic Acids Research 42. 2 (2014): 836-847. Print.
Cui, Feng, et al. "Prediction of Nucleosome Rotational Positioning in Yeast and Human Genomes Based on Sequence-dependent DNA Anisotropy." BMC Bioinformatics 15. (2014): 313. Print.
Alharbi, Bader A., et al. "nuMap: A Web Platform for Accurate Prediction of Nucleosome Positioning." Genomics Proteomics and Bioinformatics 12. 5 (2014): 249-253. Print.
Cui, F, et al. "Transcriptional Activation of Yeast Genes Disrupts Intragenic Nucleosome Phasing." Nucleic Acids Research 40. 21 (2012): 10753-10764. Print.
Macvanin, M, et al. "Noncoding RNAs Binding to the Nucleoid Protein HU in Escherichia Coli." Journal of Bacteriology 194. 22 (2012): 6046-6055. Print.
Published Conference Proceedings
Kc, Kishan, et al. "Interpretable Structured Learning with Sparse Gated Sequence Encoder for Protein-Protein Interaction Prediction." Proceedings of the 2020 25th International Conference on Pattern Recognition (ICPR), 10-15 Jan. 2021, Milan, Italy. Ed. Rita Cucchiara, Alberto Del Bimbo, and Stan Sclaroff. Milan, Italy: n.p., 2021. Web.
Li, Rui, et al. "Sparse Covariance Modeling in High Dimensions with Gaussian Processes." Proceedings of the Neural Information Processing Systems 2018. Ed. S. Bengio, et al. Montreal, Canada: n.p., 2018. Web.
Bao, Feifei, et al. "P53 Binding Sites in Normal and Cancer Cells are Characterized by Distinct Chromatin Context." Proceedings of the AACR Annual Meeting 2018. Ed. Chi Van Dang. Chicago, IL: n.p., 2018. Print.