sriram-lab / emt-cobra Goto Github PK

Use the data from this paper: https://www.nature.com/articles/s41467-021-21898-7

Points to address with high throughput datasets:

• Get a sense of robustness across different -omics modalities (CRISPR / siRNA / drug)
• Examine CRISPR batch corrected results

• CRISPR-Cas9
• Drug data
• siRNA

Previous growth rate knockout data was generated using the RECON1 metabolic reconstruction and used quartiles to delineate between up- and downregulation. This pass uses new method to generate single-cell profiles in RECON1 and iHUMAN.

Tasks:

• Debug Great Lakes software to perform constraint-based metabolic modeling
• Re-compute metabolic fluxes in RECON1
• Compute metabolic fluxes in iHUMAN
• Re-compute growth rates from gene and reaction knockout in RECON1
• Compute growth rates from gene knockout in iHUMAN

• Need to see if there is a better way to analyze time-course data
• Need to compare and contrast transcriptomics and proteomics data

Things to do:

• Create tables showing all data for bulk and single-cell experiments
• Compare and contrast early and late time points for vulnerabilities
• Identify reactions that are uniquely active in 1 phase versus another
• Add all citations
• Write abstract
• Write introduction
• Clarify language
• Discuss differences between single-cell versus bulk transcriptomics
• Disucuss differences between transcriptomics and proteomics