News

Our lab gave several presentations at CCSC-NE, which was held this year here at Smith.

In the afternoon, Halie and Jade Lili (one of our amazing CSC 210 TAs) as well as two former data structures students presented the unique data structures curriculum that Halie has been using in CSC 210 for the past few semesters. Their presentation was one of two tutorials offered at the conference this year.

In the evening, two groups from the lab also presented posters at the undergraduate research poster session:

• Ayesha & Yuzhang presented their poster “Mouse Model of Disease - Language, Metadata, and Cross-Species Comparison”

• Nargiz presented her poster “Microsatellites Made Accessible: An Open-Source Tool for Microsatellite Data Analysis”

  

  

Four student research teams submitted abstract to Intelligent Systems in Molecular Biology, one of the major computational biology conferences, which will be held in Washington D.C. in July.

Some of the projects include:

• Natural language processing approaches for identifying the methods used in a study based on its metadata (submitted to BOKR)
• Natural language processing approaches for comparing the similarity of experiments across species (submitted to BOKR)
• An open-source tool for extracting fluorescent peak information from fragment analysis files in the ABIF format (submitted to BOSC)
• An analysis of alternative clustering approaches to understanding molecular phenotypes of high-grade serous ovarian cancer (submitted to iRNA)

Notably, all of these submissions had at least one sophomore as a co-author! Fantastic work everyone!

Exciting news: the lab had two abstracts accepted at the third annual Symposium on AI in Veterinary Medicine (SAVY3.0), which will be held at Cornell University in May.

Caitlyn Kim will present her poster “A New Benchmarking Dataset of AKC-Registered Dog Breeds for Fine-Grained Image Classification Tasks”, describing the work her team has been doing to build a new dataset of images of dogs with confirmed breed labels. Co-authors include Nora and Frankie.

Elaine Chen is the lead author on the abstract “Differentiating Effects of Neural Network Architecture versus Dataset in Computer Vision’s Dog Breed Identification Problem”, which has been accepted as one of five talks in the session on AI Applications in Companion Animal Health. Co-authors include Glenvelis Perez and Yixuan (Quinn) He.

The teams are looking forward to attending this incredible symposium and sharing their work with the community!

We are excited to share that Alina Yildirim has been selected as an Honors Fellow through the AEMES McKinley Fellowship Program!

The fellowship will support Alina’s honors thesis work in the VBIL during her senior year. More information about Smith’s AEMES programs is available here.

Today Hildana Shiferaw gave an excellent presentation of her honors thesis on a urine-based screening approach for ovarian cancer!

Her honors thesis is titled “System Design of a Urine-Based Screening Approach for Ovarian Cancer Informed by Computational Identification of miRNAs as Biomarkers” and is co-advised by Halie Rando (CSC) and Mike Kinsinger (EGR). Notably, Hildana completed a programming-heavy computational biology thesis without any CSC coursework after CSC 110! We are all so proud of her achievements – congratulations Hildana!

Rising seniors Hildana Shiferaw and Tseegi Nyamdorj presented lightning talks at Computational Systems for Integrative Genomics at the Flatiron Institute in NYC!

Tseegi’s talk was titled “Manubot: Incorporating MS Word track changes into version controlled writing workflows”, and Hildana’s was titled “Comparing semantic and biological representations of animal models.”

We also got to take a tour of their in-house computing infrastructure!

We are excited to share that Hildana Shiferaw has been selected as an Honors Fellow through the AEMES McKinley Fellowship Program!

The fellowship will support Hildana’s honors thesis work in the VBIL during her senior year. More information about Smith’s AEMES programs is available here.