Software & Code

Software

• scMKL (with documentation)
  Python package for scalable multiple kernel learning using Random Fourier Features and Group Lasso. Enables scalable interpretable classification in single-cell multiomics.

• EpiConfig
  Snakemake pipeline for running the EpiConfig R package, which implements an LDA model for single-cell multiomics. Epiconfig, is an interpretable multimodal topic model that learns unsupervised clustering of single-cells while modeling cross modality relationships.

• EpiConfig Interpretation Tool
  I developed a shiny app for interpretation of multimodal EpiConfig topics to obtain biological insights into different cell states; we show that cross modality features reflect 3D genome interactions.

• UrbanOccupationsOETR (with Live app)
  Shiny app for the Industrialisation and Urban Growth from the mid-nineteenth century Ottoman Empire to Contemporary Turkey in a Comparative Perspective, 1850-2000 to help investigate the link between ethnicity and profession using spatiotemporal data.

• sgp_covid-19 (with Live App)
  R implementation of structured Gaussian processes with multiple kernel learning for the analysis of COVID-19 pandemic dynamics in the United States in Ak et al 2022.

• Prospective CCHF
  A platform for spatiotemporal analysis of Crimean-Congo Hemorrhagic Fever (CCHF). It provides access to input covariates, nationwide surveillance data, and prediction results for 2004-2017, as well as maps and model performance metrics in Ak et al 2020.

• sgp2mkl
  R implementation of structured Gaussian processes with multiple kernel learning in Ak et al. 2018.

Code

• Power analysis for spatial analysis
  Code and documentation for spatial variability and sample size effects in prostate cancer progression. Includes statistical analyses: in silico tissue generation, leave-one-out analysis, subsampling, and permutation tests in Ak et al. (2024).