Kishan KC

Ph.D. Student, RIT Computer Science

I am a second year PhD student in Computing and Information Science at the Golisano College of Computing and Information Sciences, Rochester Institute of Technology (RIT) . I am a Graduate Research Assistant under the supervision of Prof. Anne Haake and Prof. Rui Li. My research interest is to develop computational methods to provide insights into underlying biological phenomena that are critical to understanding phenotypes in health and diseases. Specifically, I am currently working in learning low-dimensional vector representation for each gene or protein, that canonically represents topological patterns in interaction networks and various information associated with that gene, which can be plugged into off-the-shelf machine learning methods for diverse functional tasks: gene function prediction, gene ontology reconstruction, and genetic interaction prediction.

I am interested in integration of heterogeneous information such as interaction networks, expression profiles, transcription factor binding sites, gene sequences, functional annotations from gene ontology, metabolic pathways, etc. to derive functional insights about genes or proteins.

Research interests: Heterogeneous Data Integration, Network Representation Learning, Deep Learning, Computational Biology


GNE: A deep learning framework for gene network inference by aggregating biological information
We present a model that integrates topological properties of gene network and gene expression data to learn representations for gene. These low-dimensional representations derive deeper insights into the structure of rapidly accumulating and diverse gene interaction networks and greatly simplify downstream modeling. Our analysis shows that these learned representations achieve significantly more accurate gene interaction predictions.
Kishan KC*, Rui Li, Feng Cui, Anne Haake


February 2018 Presented poster "Network-based Learning to Infer Genetic Interaction Networks" in New Deep Learning Techniques Workshop at IPAM, UCLA.
February 2018 Attended New Deep Learning Techniques Workshop at IPAM, UCLA.
May 2017 Defended the PhD Research Potential Assessment.
August 2016 Joined Rochester Institute of Technology as PhD Student in Computing and Information Sciences.


Gene Network Embedding
Developed a deep learning framework to integrate gene expression data with topological properties of gene interaction networks that models their relative importance in gene interaction prediction. This framework learns a lower-dimensional representations for a gene, which is used to predict its interactions with other genes.

Reconstruction of Gene Regulatory Networks with Ensemble SVM
Developed an ensemble of Support Vector machine that identifies the regulatory relationships between transcription factors and target genes. Framing the reconstruction as feature selection problem, we followed recursive feature elimination approach to identify the candidate transcription factors that are significant in expression of a target gene.

Multiplayer Web Checker
A web-based java application that allows players to play checkers with other players who are currently signed-in. The game user interface (UI) supports a game experience using drag-and-drop browser capabilities for making moves. The WebCheckers webapp uses Spark web micro framework and FreeMarker template engine to handle HTTP requests and generate HTML responses.

Knowledge Graph, a search engine
We create a search engine that searches user querying on the database of documents. The content of documents are tokenized and indexed to create a graph, representing their relationship. User query is searched against this graph to display the information.

Parallel Breadth First Search with Parallel Java 2
An implementation of Water Jug Puzzle Solver for different number of water jugs using Parallel Java Library.