Figure S1 from the paper: Main steps of gene expression
The p-gp protein is the product of the MDR1 gene. It functions as a cellular drug exporter and is known to be related to patients' response to chemotherapy. Our study focuses on understanding how genetic variants in the MDR1 gene might affect the production and function of this protein. While many non-computational studies have investigated the effects of these variants, the results have been inconsistent. Using computational tools, we aim to assess how these variants influence key steps of gene expression to understand their potential impact on patients' resistance to treatment and survival. Our findings suggest that these variants, particularly T3435C, could alter the folding and therefore function of the p-gp protein via the effect on the folding of the mRNA, potentially affecting its role in drug resistance. Additionally, we identify variants that are predicted to potentially effect mRNA and protein folding and thus may play important roles in disease. Ultimately, we propose a framework for studying the effects of genetic variants on gene expression, which could help deepen our understanding of disease mechanisms.
Figure S1 from the paper: Main steps of gene expression
This project requires three separate conda environments due to different dependencies:
Non-confidential data is available on Zenodo (https://doi.org/10.5281/zenodo.14011668) and protected data is available through the Genomic Data Commons (GDC) with appropriate access permissions.
Within the code directory there are two folders: MDR1_code and co_trans_code. MDR1_code contains the scripts for the main analyses conducted in the paper, and co_trans_code contains the co-translational folding model implementation (unpublished).
The main scripts in MDR1_code are:
expression_MDR1.ipynb: Checks variant effect on MDR1 expression of TCGA patientsEnformer_MDR1.ipynb: Predicts variant effect on MDR1 expression using EnformerSplicing_MDR1.ipynb: Predicts variant effect on MDR1 splicing using SplicAIpost_transcriptional_editing.ipynb: Predicts variant effect on A-to-I sites, m6A sites and binding sites of RNA binding proteinstranslation_rate_MDR1.ipynb: Predicts variant effect on MDR1 local translation ratesCo_translational_folding_MDR1.ipynb: Predicts whether the variants are in locations that are likely to be important for co-translational foldingsurvival_MDR1.ipynb: Predicts variant effect on patient overall and progression-free survivalmost_extreme_ClinVar.ipynb: Creates a list of the ClinVar variants that are most likely to have an effect on mRNA folding in conserved genomic positionsmost_extreme_TCGA.ipynb: Creates a list of the TCGA variants that are most likely to have an effect on mRNA folding in conserved genomic positionsThe main scripts in co_trans_code are:
retrieve_orthologs.ipynb: Retrieves all orthologs of a gene from Ensemblfrom_orthologs_to_nt_MSA.ipynb: Creates multiple sequence alignment (MSA) for the orthologs and maps to nucleotide alignmentsvertical_perm.ipynb: Creates vertical permutations of the MSA for randomizationhorizontal_perm.ipynb: Creates horizontal permutations of the MSA for randomizationmfe_windows.ipynb: Calculates the minimum free energy (MFE) of the MSA in windows of 39 positions, for both the original and permuted MSAsmfe_positions_zscores.ipynb: Obtains z-scores and p-values for the MFE of each positionValidation_zscores_clinvar.ipynb: Validates the model using ClinVar variantsValidation_zscores_TCGA.ipynb: Validates the model using TCGA variantsThis code is freely available for academic use. For non-academic or commercial usage, please contact the authors.
If you use this code in your research, please cite:
Gutman, T., Tuller, T. Computational Analysis of MDR1 Variants Predicts Effect on Cancer Cells via their Effect on mRNA Folding. [Manuscript in preparation]
For questions or issues, please contact: