Compositions and methods for treating ras-mutant cancers
A technology for mutants and cancers, applied in the field of compositions and methods for treating RAS mutant cancers, capable of solving problems such as overstimulation of signal transduction pathways
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0208] Embodiment 1: experimental material and method
[0209] Preprocessing of scRNA-seq count data. UMI count matrices of raw HGNC alignments generated via 10× sequencing were preprocessed and scaled before analysis in the downstream analysis pipeline. Low-abundance genes (eg, mean count <0.25) and genes with reads in <10% of cells and cells with non-zero reads for <10% of all genes were removed from the count matrix. To adjust for differences in sequencing depth between individual cells, count matrices were in some cases normalized and scaled prior to subsequent analysis. Methods of normalization include, but are not limited to: 1) globally scaling cell-level counts to match median depth across all cells (scalar adjustment) and 2) solving the linear system to obtain unique scaling factors for individual cells. In some cases, sample-batch effects were corrected via a mutual nearest neighbor algorithm.
[0210] Supervised dimensionality reduction. To computationally ide...
Embodiment 2
[0226] Example 2: Computational discovery and validation of novel targets in Kras mutant pancreatic cancer
[0227] Figures 1A-1B provide a pipeline for computational discovery of therapeutic targets in RAS mutant pancreatic cancer. Kras mutant pancreatic cancer cell lines were screened using the CRISPRi library transduced at a low multiplicity of infection (MOI) as shown in Figure 1A. Transcriptomes of single cells were isolated and converted into DNA libraries using Chromium Instruments (10X Genomics) and enzyme kits (10X Genomics) and sequenced using the Hiseq4000 system (Illumina). Single-cell RNA sequencing (scRNA-seq) profiles of individual cells were mapped to their respective CRISPR targets via paired-end sequencing and using barcodes. Raw reads in FASTQ format were aligned to the whole genome and mapped to the appropriate genomic coordinates and HGNC genes for each gene, resulting in a count matrix consisting of N cells × M genes (see Figure 1A). The N×M matrix is...
Embodiment 3
[0229] Example 3: Computational identification of TXNRD1 as a therapeutic target in KRAS mutant pancreatic cancer
[0230] As shown in Figure 1B, scRNA-seq profiles of individual Kras mutant PDAC cells expressing CRISPRi targets were analyzed using the decision function of a machine learning algorithm trained on two off-target guide RNAs, a toxic target, and a healthy ductal cell line as described above . The output of the decision function is further transformed via various methods including but not limited to: effect size, K.S. statistics, z-score or p-value relative to a control population. Scores across multiple machine learning algorithms and repeated experiments were further aggregated via various methods including, but not limited to: average, weighted average, rank aggregation, weighted rank aggregation, Stouffer's method (z-score), and Fischer's method (p-value).
[0231] As shown in Figure 2A, TXNRD1 was identified as a CRISPRi target that transformed Kras mutant...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More