Systems and methods for tracking the evolution of single cells

JP2023518416A5Active Publication Date: 2024-12-04GENENTECH INC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
JP2022556040
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2021-03-15
Filing Date
2021-03-19
Publication Date
2024-12-04
Estimated Expiration
2041-03-19

AI Technical Summary

Technical Problem

Cancer cells often develop resistance to therapy, leading to recurrent tumors, and current methods struggle to understand the mechanisms of resistance to various treatments, hindering the development of effective treatment combinations.

Method used

A method involving barcoded cells is used to track single-cell clones over time, analyzing their fitness and transcriptional status under selective pressure, allowing for the identification of pre-existing traits and adaptive changes that confer resistance or sensitivity to treatments, using single-cell RNA sequencing (scRNA-seq) to compare 'winner' and 'loser' clones.

Benefits of technology

Enables the identification of cellular programs that facilitate adaptation to therapeutic pressures, providing insights into drug resistance and response mechanisms, enabling more effective treatment strategies by capturing heterogeneity at the single-cell level.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

Cells in a given population often exhibit heterogeneity, which can affect how each cell responds to a particular treatment or growth condition. The methods described herein allow for determining which cells from an initial population survive a treatment or condition, and how surviving cells evolve over time. For example, the methods described herein can be used to model drug resistance, response, and / or adaptation in a cell population.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical Field]

[0001] Cross-reference of related applications This application claims priority to U.S. Provisional Application No. 62 / 992,498, filed on 20 March 2020, and U.S. Provisional Application No. 63 / 161,390, filed on 15 March 2021, both of which are incorporated herein by reference. [Background technology]

[0002] background In recent years, there have been remarkable advances in cancer treatment, but tumor cells often develop resistance to treatment, thereby escaping death. These resistant cells then disseminate recurrent tumors that require secondary treatment, if available. Co-treatment with multiple therapies has been clinically shown to provide further benefits to patients by partially overcoming resistance. Understanding the mechanisms of resistance to various treatments is desirable in order to develop more efficient treatments and combinations of treatments for cancer. [Overview of the project] [Means for solving the problem]

[0003] overview Cells in a given population often exhibit heterogeneity that can influence how each cell responds to a particular treatment or growth condition. The methods described herein enable the determination of which cells from the initial population will survive the treatment or condition, and how such cells survive and / or evolve the treatment over time. For example, the methods described herein can be used to model drug resistance, response, and / or adaptation in cell populations.

[0004] This technology generally enables the tracking of single-cell clones over time and the comparison of their suitability under desired selective pressures. The suitability of each single-cell clone can be determined by its relative size over time, which can be measured, for example, by the abundance of a unique barcode genetically introduced into that clone. Simultaneously, each single-cell clone can be analyzed, for example, by single-cell RNA sequencing (scRNA-seq) to obtain the transcriptional state at single-cell resolution before, during, and after selection. By comparing the transcriptional programs associated with "winner" clones (cells with enriched barcodes) with those of "loser" clones (cells with depleted barcodes), it is possible to identify existing traits as well as adaptive changes that enable survival under desired selective pressures, such as drug treatment, genomic engineering, and engraftment in a host.

[0005] In one embodiment, a method is provided for screening cells for a trait. This method may include: (a) obtaining a plurality of barcoded cells, each containing a single unique barcode or combination of barcodes; (b) performing a first sequencing of RNA and / or DNA on a subset of the plurality of barcoded cells; (c) culturing the plurality of barcoded cells for a first period in the presence of selective pressure to form a first plurality of cells; (d) performing a second sequencing of RNA and / or DNA on a subset of the first plurality of cells; (e) culturing the first plurality of cells for a second period in the presence of selective pressure to form a second plurality of cells; and (f) performing a third sequencing of RNA and / or DNA on at least a subset of the second plurality of cells. This method may also include: (g) determining the relative abundance of the barcodes sequenced in the first sequencing, the second sequencing, and / or the third sequencing.

[0006] In another embodiment, a method is provided for screening cells for response traits to a therapeutic agent. This method may include: (a) obtaining a plurality of barcoded cells, each containing a single unique barcode; (b) performing a first sequencing of RNA and / or DNA on a subset of the plurality of barcoded cells; (c) culturing the plurality of barcoded cells for a first period in the presence of the therapeutic agent to form a first plurality of cells; (d) performing a second sequencing of RNA and / or DNA on a subset of the first plurality of cells; (e) culturing the first plurality of cells with the therapeutic agent for a second period to form a second plurality of cells; and (f) performing a third sequencing of RNA and / or DNA on a subset of the second plurality of cells. This method may also include: (g) determining the relative abundance of the barcodes sequenced in the first sequencing, the second sequencing, and / or the third sequencing.

[0007] In another embodiment, a method is provided for comparing responses to selective pressure. This method may include: (a) obtaining a first plurality of barcoded cells, each barcoded cell having a single unique barcode; (b) obtaining a second plurality of barcoded cells substantially similar to the first plurality of barcoded cells; (c) performing a first sequencing of RNA and / or DNA from the first plurality of barcoded cells and / or the second plurality of barcoded cells; (d) culturing the first plurality of barcoded cells in the presence of a first selective pressure to form the first plurality of cells; (e) culturing the second plurality of cells in the presence of a second selective pressure (f) a step of culturing barcoded cells to form a second group of cells; (g) a step of performing a second sequencing of RNA and / or DNA from the first group of cells and / or the second group of cells; (h) a step of culturing the first group of cells in the presence of a first selective pressure to form a third group of cells; (i) a step of culturing the second group of cells in the presence of a second selective pressure to form a fourth group of cells; and (j) a step of performing a third sequencing of RNA and / or DNA from the third group of cells and / or the fourth group of cells. The method may also include: (j) a step of determining the relative abundance of one or more barcodes sequenced in the first sequencing, the second sequencing, and / or the third sequencing.

[0008] In one embodiment, a method is provided for screening cells for phenotypes in the cells. This method may include: (a) providing a mixture of cells comprising a plurality of clonal populations, wherein each clonal population comprises an identifier unique to that clonal population, and the initial genetic, transcriptome, and / or proteome information of at least one representative member of each clonal population is known; (b) culturing the mixture of cells in the presence of a first selective pressure for a first period, and at the end of the first period, obtaining second genetic, transcriptome, and / or proteome information of at least one member of a surviving clonal population from the mixture of cells; (c) subjecting the mixture of cells subjected to the first selective pressure to a second selective pressure for a second period, and at the end of the second period, obtaining third genetic, transcriptome, and / or proteome information of at least one member of a surviving clonal population from the mixture of cells; and (d) determining the relative abundance of each clonal population present in the final mixture of cells based on the unique identifier relating to the clonal population. This method may also include (e) a step of identifying adaptive traits, which are genetic and / or proteomic traits that are present or absent in the clonal population in the final mixture of cells. This method may also include (f) comparing information from the first genetic, transcriptome, and / or proteomic information, the second genetic, transcriptome, and / or proteomic information, and / or the third genetic, transcriptome, and / or proteomic information, and / or the one or more subsequent genetic, transcriptome, and / or proteomic information.This method may also include (g) comparing the initial genetic information, transcriptome information, and / or proteome information, the second genetic information, transcriptome information, and / or proteome information, and / or the third genetic information, transcriptome information, and / or proteome information, and / or the one or more subsequent genetic information, transcriptome information, and / or proteome information with the genetic information, transcriptome information, and / or proteome information of different clonal populations of cells having different unique barcodes that have been subjected to selective pressure.

[0009] In another embodiment, a method is provided for identifying cellular programs that facilitate adaptation to pressure. This method may include: (a) transducing cells with multiple barcodes such that each cell contains a single unique barcode or combination of barcodes; (b) growing the cells in culture to produce a starting cell pool of clones of cells containing each barcode; (c) obtaining first genetic information, transcriptome information, and / or proteome information from a first subset of the starting cell pool; and (d) growing a second subset of the starting cell pool in the presence of selective pressure to grow the starting cell pool and form an intermediate cell pool. The method includes: (e) obtaining second genetic information, transcriptome information, and / or proteome information from a first subset of an intermediate cell pool; (f) continuing to culture the second subset of the intermediate cell pool in the presence of selective pressure to grow the intermediate cell pool and form a final cell pool; (g) obtaining third genetic information, transcriptome information, and / or proteome information from at least a subset of the final cell pool; and (h) quantifying the level of each barcode in the final cell pool, intermediate cell pool, and / or starting cell pool. The method may also include: (i) assigning cells having concentrated barcodes in the final cell pool as winning clones, and / or assigning cells having depleted barcodes in the final cell pool as losing clones. The method may also include: (j) determining gene mutations, transcriptional programs, and / or protein expression associated with at least one winning clone and / or at least one losing clone. [Brief explanation of the drawing]

[0010] [Figure 1] This schematic diagram illustrates an exemplary overview of longitudinal transcriptomics of barcoded cells (bottom), combining conventional endpoint studies (top left) with clonal analysis (top right) to enable the analysis of the response and adaptation of individual cells / clones to selective pressure over time.

[0011] [Figure 2]Schematic showing an example of expression-based high-complexity barcoding combined with single-cell analysis for the analysis of sensitivity and resistance to drug treatment (or other selection pressures) over time.

[0012] [Figure 3] Exemplary scRNA-seq compatible barcoding construct with barcode insertion site, T2A (self-cleaving peptide), and labeled EGFP and puromycin selection gene is shown. Insertion of the construct into cells results in the expression of EGFP transcripts containing the barcode. The EGFP-barcode transcripts can be sequenced by RNA-seq.

[0013] [Figure 4-1] Figures 4A and 4B are graphs showing the abundance (as log2) of each barcode at day 4 (Y-axis) compared to the endpoint (X-axis) of treatment with erlotinib (Figure 4A) or the degrader (Figure 4B). Barcodes associated with high levels of abundance at both time points are labeled. Points corresponding to the selected barcodes are labeled. [Figure 4-2] Same as above.

[0014] [Figure 5] Graph from Figure 4A, with clones predicted to survive (left circle) and clones predicted to survive and adapt to erlotinib (right circle) circled.

[0015] <0​​​​​​​​​​​ [Figure 8] This is a heatmap of sample barcode abundances before and after erlotinib treatment (left). The barcodes were classified into resistance and susceptibility classes based on their relative abundance changes. Differential gene expression analysis was performed to determine the genes differentially expressed between the two barcode classes enclosed in boxes. Then, pathway enrichment analysis was performed using the differential expression gene list to determine the pathways and their gene members in which resistance clones were statistically significantly enriched (right).

[0018] [Figure 9] (Left) A heatmap of barcode abundance, where barcodes were clustered based on relative abundance changes after erlotinib and EGFR degrader treatment. Differential gene expression and pathway enrichment analysis between barcode classes enclosed in two boxes identified statistically significant enriched pathways and their gene members in EGFR degrader-sensitive clones (right).

[0019] [Figure 10-1] (Left) A heatmap of barcode abundance, where barcodes were clustered based on the change in relative abundance after treatment with erlotinib and EGFR degrader. Barcodes showing differential responses to erlotinib versus EGFR degrader (box barcodes, degrader resistant but with much lower erlotinib resistance) were selected for differential gene expression analysis to compare the transcriptional responses of these cells to treatment with erlotinib or degrader, and gene-related pathways were identified (right).

[0020] [Figure 10-2] A violin plot (left) showing ATF4 and SLC7A5 in untreated cells, erlotinib-treated cells, and degrader-treated cells, and the potential mechanism of action of erlotinib in degrader-resistant clones (right).

[0021] [Figure 11-1]This document outlines an experimental strategy for determining the effect of EGFR protein depletion on erlotinib efficacy using PC9 cells.

[0022] [Figure 11-2] This bar graph shows the relative gene expression of integrated stress response genes and MAPK genes in PC9 cells treated with siNTC + erlotinib (100 nM), siEGFR + DMSO, or siEGFR + erlotinib (100 nM). Gene expression is normalized relative to siNTC + DMSO. The legend from top to bottom represents each of the three groups of bars from left to right.

[0023] [Figure 11-3] This is a Violin plot showing MAPK gene expression in untreated cells (day 1), erlotinib-treated cells (day 5), or degrader-treated cells (day 5).

[0024] [Figure 11-4] Images of PC9 cells stained with crystal violet are shown to evaluate cell viability after treatment with siNTC + erlotinib (100 nM), siEGFR + DMSO, or siEGFR + erlotinib (100 nM).

[0025] [Figure 12] Images of H1975 cells stained with crystal violet to evaluate cell viability after treatment with 100 nM osimertinib (left), 100 nM osimertinib and 1 μM control compound (center), or 100 nM osimertinib + 1 μM degrader (right).

[0026] [Figure 13-1] This is a schematic diagram showing pharmacological modulo agents of ER stress (Tg: thapsigargine; Tm: tunicamycin) that lead to the induction of the integrated stress response (ISR). Integrated stress response inhibitors (ISRIBs) inhibit pelF2α, and as a result inhibit the ISR.

[0027] [Figure 13-2] The bar graph (left) shows the percentage survival rate of cells treated with a degrader, degrader + Tm, or degrader + Tg, and the image (right) shows cells stained with crystal violet to evaluate cell viability at the time of treatment.

[0028] [Figure 13-3] The bar graph (left) shows the percentage survival rate of cells treated with degrader, erlotinib (Erl), or Erl+ISRIB, and the image (right) shows cells stained with crystal violet to assess cell viability at the time of treatment.

[0029] [Figure 14] This is a heatmap of cell clone abundance measured by NGS analysis of barcodes. Blue boxes indicate degrader-resistant clones or erlotinib-resistant clones.

[0030] [Figure 15] This bar graph shows the expression of ISR genes and MAPK target genes (compared to control) as measured by qPCR in cells treated with erlotinib, erlotinib and ISRIB, degrader, or ISRIB alone.

[0031] [Figure 16] This is a schematic diagram illustrating the mechanistic pathway for verifying the effectiveness of EGFR kinase inhibitors.

[0032] [Figure 17-1]TraCe-seq reveals the loss of efficacy of dual EGFR inhibitor-degrader compared to conventional EGFR kinase inhibitors. Figure 17A shows a schematic diagram of the TraCe-seq workflow that enables clonal tracking and compatibility mapping as a function of time. This allows for longitudinal evaluation of gene expression and direct comparison of the mechanisms of efficacy and response or resistance to different treatments at single-cell resolution. Figure 17B shows a schematic diagram of the TraCe-seq barcoded construct with an 8nt sublibrary index and a 30nt GC-optimized barcode. Figure 17C shows the chemical structures of erlotinib, erlotinib-derived EGFR degrader GNE-104, and non-degrader control GNE-069. Figure 17D shows a schematic diagram of a TraCe-seq experimental design for comparing the mechanisms of efficacy and response and resistance to EGFR kinase inhibitors (erlotinib and GNE-069) versus EGFR degrader (GNE-104). Figure 17E shows the number of cells captured by single-cell RNA sequencing under each condition. Figure 17E is a bar graph showing TraCe-seq barcode diversity (Shannon index) before (baseline) and under treatment. The number of TraCe-seq barcodes recovered from single-cell RNA-seq reads is shown for each bar. Figure 17F is a violin plot showing the expression of MAPK pathway signature genes obtained from single-cell RNA-seq before (baseline) and under treatment (*p<10-100). Figure 17G is a bar graph showing the percentage of cells in the G0 cell cycle state before (baseline) and under treatment, as inferred from gene expression. [Figure 17-2] Same as above. [Figure 17-3] Same as above. [Figure 17-4] Same as above. [Figure 17-5] Same as above. [Figure 17-6] Same as above. [Figure 17-7] Same as above.

[0033] [Figure 18-1]TraCe-seq reveals different transcriptional states associated with the response to and resistance of erlotinib versus the erlotinib-derived degrader GNE-104. Figure 18A shows a heatmap of treatment-resistant clones showing their relative abundances before (baseline) and under treatment with erlotinib, GNE-069, or GNE-104. Figure 18B shows a violin plot showing normalized pre-existing expression levels of VIM and AXL in cells belonging to the TraCe-seq barcode category before drug treatment. Figure 18C shows a gene set enrichment analysis showing that protein processing in the ER gene is significantly depleted in degrader-resistant cells compared to degrader-sensitive cells before treatment. Figure 18D shows UMAP visualizations of all cells treated with erlotinib, GNE-069, and GNE-104. Arrow lines indicate trajectories derived by Slingshot. Figure 18E shows a schematic diagram showing the relationship between UMAP clusters and trajectories inferred by Slingshot. Figure 18F shows the trends of various pathway expression outputs plotted as a function of spurious time along each trajectory. Figure 18G shows a bar graph showing the normalized distribution of kinase inhibitor-resistant cells versus degrader-resistant cells and end clusters for each trajectory. Figure 18H shows a gene enrichment analysis showing that protein processing in the ER gene is significantly upregulated under erlotinib treatment compared to GNE-104 treatment in degrader-resistant cells. [Figure 18-2] Same as above. [Figure 18-3] Same as above. [Figure 18-4] Same as above. [Figure 18-5] Same as above. [Figure 18-6] Same as above. [Figure 18-7] Same as above. [Figure 18-8] Same as above.

[0034] [Figure 19-1]Validation of TraCe-seq predictions reveals that activation of EGFR protein and integrated stress response genes downstream of ER stress contributes to the cytotoxic activity of EGFR kinase inhibitors. Figure 19A shows a clonal assay demonstrating that siEGFR promoted PC9 cell survival under treatment with the EGFR kinase inhibitors erlotinib and osimertinib. Figure 19B shows a clonal assay demonstrating that the allosteric degrader GNE-641 promoted NCI-H1975 cell survival under osimertinib treatment, while the non-degrader control GNE-640 did not. EGFR and pEGFR levels were analyzed by Western blotting under the indicated treatment conditions. Cells were pre-treated with osimertinib for 4 hours to ensure near saturation of EGFR protein with this covalent inhibitor, followed by co-treatment with GNE-640 or GNE-641 for 24 hours. Figure 19C shows a schematic model illustrating how inhibitor-bound EGFR protein contributes to the cytotoxic activity of EGFR kinase inhibitors and how EGFR degradation attenuates such activity. Figures 19D and 19E show qRT-PCR quantification of key downstream pro-death genes of ER stress in PC9 and NCI-H1975 cells treated for 3 days under the indicated conditions. Figure 19F shows a clonal assay demonstrating that co-treatment with ISRIB promoted the survival of PC9 cells under erlotinib treatment. Figures 19G and 19H show clonal assays demonstrating that the PERK activator CCT020312 significantly enhanced the potency of FDA-approved EGFR inhibitors erlotinib and osimertinib in PC9 (19G) and NCI-H1975 (19H) cells. Figure 19I shows a bar graph comparing proliferation inhibition by EGFRi + CCT 020312 compared to monotherapy. All error bars represent SD (Score). [Figure 19-2] Same as above. [Figure 19-3] Same as above. [Figure 19-4] Same as above. [Figure 19-5] Same as above. [Figure 19-6] Same as above. [Figure 19-7]Same as above. [Figure 19-8] Same as above. [Figure 19-9] Same as above.

[0035] [Figure 20-1] Quality control metrics for the recovery and assignment of TraCe-seq barcodes. Figure 20A shows single-cell RNA-seq results obtained from a mixture of five cell lines, each labeled with five different TraCe-seq barcodes, visualized using UMAP. Clustering was performed based on transcriptome differences between cells (left side, UMAP plot labeled by transcriptome clusters) and annotated by assigned TraCe-seq barcodes (right side, UMAP plot labeled by TraCe-seq barcodes). As expected, the majority of cells with a given TraCe-seq barcode label corresponded to a specific cell line / cluster. Figure 20B shows heatmaps of the top three marker genes from each cluster based on TraCe-seq annotation in bulk RNA-seq (left panel) and scRNA (right panel), supporting the idea that TraCe-seq labeling can recover key transcriptional features of each barcoded population. Figure 20C shows a violin plot showing marker gene expression of individual genes in each cell line in the mixture. The left panel of Figure 20D shows the FACS enrichment of the top 50% of eGFP-expressing cells. The four panels on the right show the loss rates of TraCe-seq barcodes compared to endogenous gene expression before and after FACS sorting in NCI-H1373 and PC9 cells. In these plots, red dots represent TraCe-seq barcodes, and blue dots represent endogenous gene groups with the same median loss rate for endogenous genes. Figure 20E is a box plot showing that cells with misassigned TraCe-seq barcodes had significantly lower barcode expression levels. The bottom of the box represents the first quartile, and the top represents the third quartile. The line represents the median. [Figure 20-2] Same as above. [Figure 20-3] Same as above. [Figure 20-4] Same as above. [Figure 20-5] Same as above.

[0036] [Figure 21-1] Figure 21A shows a Western blot illustrating dose-dependent EGFR degradation induced by GNE-104 in HCC827 cells. Figure 21B is a Western blot showing the effects of GNE-104 versus GNE-069 on EGFR and pEGFR in HCC87 cells. Figure 21C is a Western blot showing that excess free VHL ligand inhibits GNE-104-induced EGFR proteolysis in HCC827 cells, but no effect was observed for GNE-069. Figure 21D is a characterization of the biochemical potency of GNE-104 and GNE-069 (against a panel of 218 kinases) in an in vitro kinase inhibition assay. [Figure 21-2] Same as above. [Figure 21-3] Same as above. [Figure 21-4] Same as above.

[0037] [Figure 22-1] Characterization of the response of PC9 cells to erlotinib, GNE-069, and GNE-104. Figure 22A shows a dendogram of the TraCe-seq barcode enrichment patterns in replicates across 500 different PC9 clones treated with either erlotinib (2 μM) or GNE (1 μM) over a two-month period. The barcode enrichment was highly reproducible in individual treatments and differed between erlotinib treatment and GNE-104 treatment. Figures 22B–22D compare the antiproliferative effects of erlotinib, GNE-069, and GNE-104 in PC9 cells by clonality assay (22B), cell counting (22C), and incsite imaging (22D). All three compounds were applied at 1 μM. Erlotinib and GNE-069 showed comparable activity by all three measurements, while GNE-104 was less effective. [Figure 22-2] Same as above. [Figure 22-3] Same as above. [Figure 22-4] Same as above.

[0038] [Figure 23-1] TraCe-seq barcode enrichment and depletion analysis. Figure 23A shows that deep NGS sequencing of TraCe-seq clone abundance (x axis) derived from genomic DNA correlates with TraCe-seq clone abundance (y axis) derived from scRNA-seq. The cutoff (line) shows the minimum TraCe-seq barcode for downstream clone depletion analysis shown in Figure 23D. Figure 23B shows a comparison of TraCe-seq clone abundance under different treatments. Points are the number of cells with a given TraCe-seq barcode under the indicated conditions. The Pearson correlation coefficient of the barcode abundance distribution is shown. Figure 23C shows the distribution of log2 factor change in TraCe-seq barcodes under each treatment condition compared to baseline. Figure 23D is a heatmap showing the relative abundance of depleted TraCe-seq barcodes upon erlotinib, GNE-069, or GNE-104 treatment. [Figure 23-2] Same as above. [Figure 23-3] Same as above. [Figure 23-4] Same as above.

[0039] [Figure 24-1]Figure 24A shows a clonality assay confirming the differential activity of an EGFR kinase inhibitor against degraded GNE-104. Figure 24A also shows a clonality assay demonstrating the different antiproliferative effects of GNE-104 compared to erlotinib or GNE-069 across four different EGFR-mutated lung cancer cell lines. Figure 24B shows quantification of the relative survival of the same four EGFR-mutated lung cancer cell lines shown in panel (a) over 14 days under treatment with degraded GNE-104 or non-degraded control GNE-069 against erlotinib, using a CellTiter-Glo luminescent cell viability assay. Figure 24C shows a clonality assay demonstrating that high concentrations of free VHL ligand (10 μM) did not affect the cellular response to erlotinib or GNE-069 in PC9 cells. Inactive VHL enantiomers were induced as a further control. [Figure 24-2] Same as above. [Figure 24-3] Same as above.

[0040] [Figure 25-1] Abundance of treatment resistance against treatment-sensitive TraCe-seq barcodes in the estimated trajectories. Figure 25A is a density plot showing the distribution of cells with kinase inhibitor resistance versus kinase inhibitor sensitivity TraCe-seq barcodes in the UMAP space treated with erlotinib / GNE-069. Each gray dot represents an individual cell. Figure 25B shows a comparison of the resistance vs. sensitivity barcode category distribution in the terminal UMAP clusters of the four inferred trajectories. Pathways a, b, and c are each dominated by at least one resistance clone category (presumably representing adaptation / resistance), while path d is presumed to represent response due to the highest relative abundance of drug-sensitive clones. Figure 25C is a density plot showing the distribution of cells with degrader resistance barcodes subjected to GNE-104 treatment against treatment with the kinase inhibitor erlotinib / GNE-069. Each gray dot represents an individual cell. [Figure 25-2] Same as above. [Figure 25-3] Same as above.

[0041] [Figure 26-1] Further characterization of cells treated with siEGFR and kinase inhibitors or osimertinib + allosteric EGFR degrader. Figure 26A shows a schematic diagram of the siEGFR experimental setup in PC9 and HCC4006 cells. Figure 26B shows Western blot analysis of total EGFR and pEGFR after 2 days of the indicated treatment. Figure 26C shows qRT-PCR analysis of key MAPK transcription targets in PC9 cells treated for 3 days under the indicated conditions. Figure 26D shows qRT-PCT analysis of key transcription targets of the MAPK pathway in HCC4006 cells under the indicated conditions for 5 days. Figure 26E shows a clonality assay demonstrating that siEGFR promoted the survival of HCC4006 cells under treatment with the EGFR kinase inhibitors erlotinib and osimertinib. Figure 26F shows qRT-PCR analysis of key death-promoting genes downstream of ER stress in HCC4006 cells under the indicated conditions on day 5. Figure 26G shows qRT-PCR analysis of key transcriptional targets of the MAPK pathway in NCI-H1975 cells under the indicated conditions on day 3. ND: Not detected. All error bars represent SD. [Figure 26-2] Same as above. [Figure 26-3] Same as above. [Figure 26-4] Same as above. [Figure 26-5] Same as above. [Figure 26-6] Same as above. [Figure 26-7] Same as above.

[0042] [Figure 27-1]Characterization of GNE-640 and GNE-641. Figure 27A shows the chemical structures of allosteric EGFR ligands EAI-045, GNE-641 (active degrader), and GNE-640 (inactive degrader). Figure 27B is a Western blot showing that the addition of free VHL ligand can rescue EGFR degradation upon GNE-641 treatment in NCI-H1975 cells. GNE-640 was included as a further control. Figure 27C shows a clonality assay demonstrating the very modest monotherapy activity of EAI-045, GNE-640, or GNE-641 in NCI-H1975 cells compared to osimertinib. Figure 27D is a quantification showing the relative viability of NCI-H1975 cells co-treated with 0.1 μM osimertinib alone for 14 days using the CellTiter-Glo luminescent cell viability assay. Survival rates are normalized for osimertinib only. Figure 27E shows a clonality assay, demonstrating that the allosteric degrader GNE-641 promoted the survival of NCI-H3255 cells under osimertinib treatment, while the degrader control GNE-640 did not. [Figure 27-2] Same as above. [Figure 27-3] Same as above. [Figure 27-4] Same as above. [Figure 27-5] Same as above.

[0043] [Figure 28-1]Pharmacological modulation of ER stress alters the response to EGFR kinase inhibitors and degraders. Figure 28A is a schematic diagram showing pharmacological modifiers of the ER stress pathway. Figure 28B is qRT-PCR quantification of erlotinib downstream of key transcriptional targets of the ER stress and MAPK pathway in PC9 cells under the indicated treatment conditions on day 4. Figure 28C is a clonality assay showing the attenuation of osimertinib activity by ISRIB in NCI-H1975 cells. Figure 28D is qRT-PCR quantification of erlotinib downstream of key transcriptional targets of the ER stress and MAPK pathway in NCI-H1975 cells under the indicated treatment conditions on day 3. ND: Not detected. Figure 28F is a quantitative comparison of the efficacy changes of erlotinib compared to GNE-104 co-treated with the indicated pharmacological ER stress modifiers in PC9 cells using a CellTiter-Glo luminescent cell viability assay on day 5. Figure 28G shows qRT-PCR quantification of key death-promoting genes downstream of ER stress in PC9 cells treated under the indicated conditions on day 3. All error bars are SD. [Figure 28-2] Same as above. [Figure 28-3] Same as above. [Figure 28-4] Same as above. [Figure 28-5] Same as above. [Figure 28-6] Same as above. [Figure 28-7] Same as above.

[0044] [Figure 29-1]The PERK activator CCT020312 enhances the activity of FDA-approved EGFR kinase inhibitors. Figures 29A–29C show qRT-PCR quantification of ER stress and downstream cell death-promoting genes of MAPK target genes in PC9 cells (29A–29B) and NCI-H1975 cells (29C) under indicated conditions on day 3. Figure 29D shows quantification using the caspase-Glo3 / 7 assay system, demonstrating that the EGFRi + CCT020312 combination was more effective in inducing cell death compared to the single agent. All error bars represent SD. [Figure 29-2] Same as above. [Figure 29-3] Same as above. [Figure 29-4] Same as above. [Modes for carrying out the invention]

[0045] Detailed explanation After reading this description, it will become clear to those skilled in the art how to implement the disclosure in various alternative embodiments and applications. However, not all of the various embodiments of the invention are described herein. It will be understood that the embodiments presented herein are presented only as examples and are not limiting. Accordingly, this detailed description of various alternative embodiments should not be construed as limiting the scope or breadth of the disclosure as described herein.

[0046] Before disclosing and describing this technology, it should be understood that the embodiments described below are not limited to specific compositions, methods for preparing such compositions, or uses thereof, and are therefore naturally subject to change. It should also be understood that the terminology used herein is intended solely to describe specific embodiments and is not intended to limit them.

[0047] Detailed explanations divided into various sections for the convenience of the reader, and disclosures found in any section, may be combined with those in other sections. Titles or subtitles may be used herein for the convenience of the reader and are not intended to affect the scope of this disclosure.

[0048] definition Unless otherwise defined, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which this disclosure pertains. In this specification and the following claims, several terms are referenced that must be defined as having the following meanings.

[0049] The terms used herein are intended solely to describe specific embodiments and are not intended to limit them. As used herein, the singular forms “a,” “an,” and “the” are intended to include the plural forms as well, unless the context otherwise explicitly indicates otherwise.

[0050] "Optional" or "optionally" means that the event or situation described below may or may not occur, and that the description includes both the cases in which the event or situation occurs and the cases in which it does not.

[0051] The term "approximately," when used before a numerical specification that includes a range, such as temperature, time, quantity, or concentration, indicates an approximate value that may vary by (+) or (-) 10%, 5%, 1%, or any sub-range or sub-value between them. Preferably, when used in relation to quantity, the term "approximately" means that the quantity may vary by + / - 10%.

[0052] The terms “comprising” or “comprises” are intended to mean that a composition and method includes the elements described but does not exclude others. “Consisting essentially of,” when used to define a composition and method, means excluding other elements that are essentially important to the combination for the described purpose. Thus, a composition consisting essentially of the elements defined herein does not exclude other materials or processes that do not substantially affect the basic and novel features (one or more) of the claimed invention. “Consisting of” means excluding other components and elements that are more than trace amounts of substantial method processes. Embodiments defined by each of these transitional terms are within the scope of this disclosure.

[0053] A “barcode” refers to one or more nucleotide sequences used to identify a cell or clonal population to which a barcode is associated. A barcode can be 3 to 1000 or more nucleotides long, preferably 10 to 250 nucleotides, more preferably 10 to 30 nucleotides, and includes any length within these ranges, for example, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 40, 50, 60, 70, 80, 90, 100, 200, 300, 400, 500, 600, 700, 800, 900, or 1000 nucleotides. A barcode is “unique” if (statistically) the barcode is found in about one cell within a population of cells. Next, cells containing barcodes can be propagated to create a clonal population, such that each cell in the clonal population contains the same barcode. For example, "multiple barcoded cells, each containing a single unique barcode" may refer to a population of cells that (statistically) contain a single cell containing a given barcode or a unique combination of barcodes. Alternatively, it may refer to a population of cells that contains multiple clonal populations of cells, where each cell in each clonal population contains the same barcode, but cells in different clonal populations contain different barcodes.

[0054] The terms “sensitive” or “sensitivity” are used herein to refer to the responsiveness of a cell or cell population to a selective pressure or therapeutic agent. Cellular responsiveness may be growth arrest, quiescence, senescence, apoptosis, or other forms of programmed cell death. Cellular responsiveness may be an intended response to a selective pressure or therapeutic agent, such as apoptosis in response to a cytotoxic agent. Cellular responsiveness may be a change in cellular properties induced by a selective pressure, including cell fate. Cellular responsiveness may be cell plasticity, cell reprogramming, growth dynamics, or metastatic ability. In embodiments, the selective pressure may be treatment with a therapeutic agent, contact with a contaminant, genomic engineering, engraftment in a host, culture conditions, growth conditions, contact with a stimulus, or contact with other cells. In embodiments, the therapeutic agent may be a cancer drug (e.g., a kinase inhibitor or other chemotherapeutic agent).

[0055] The term "resistance" is used herein to refer to the absence of an intended response by a cell or cell population to a selective pressure or therapeutic agent. In embodiments, resistance may be adaptation to a selective pressure or therapeutic agent. In embodiments, resistance may be due to one or more pre-existing characteristics of a cell or cell population. In embodiments, resistance may be acquired, for example, by activation of survival pathways in response to a selective pressure or therapeutic agent. For example, resistance may include cancer cells adapting to cancer treatment and surviving.

[0056] It will be understood that the examples and embodiments described herein are for illustrative purposes only, and that various modifications or changes taking them into account should be proposed to those skilled in the art and should be included within the spirit and scope of this application and the appended claims. All publications, patents, and patent applications cited herein are incorporated herein in their entirety by reference for all purposes.

[0057] method While not bound by theory, the methods described herein are expected to capture heterogeneity in cell populations with clonal granularity. The methods described herein may enable simultaneous monitoring of susceptibility and resistance clonal responses to treatments or conditions, and / or identification of existing features that confer resistance without the need to establish fully resistant cells. Furthermore, the methods described herein may enable comparison of mechanisms of action (MOAs) of different molecular entities in heterogeneous populations at clonal cell resolution, and / or tracking of cellular evolutionary trajectories. In addition to oncology, these methods may be applied to other disease areas for characterizing desired cellular evolutionary processes, including but not limited to cell reprogramming and cell engineering.

[0058] For example, multiple cells, each bearing a label (e.g., a barcode) to identify a cell clone or clonal population, are sequenced at a first point in time before being subjected to selective pressure. The cells are then subjected to selective pressure for a certain period, after which the surviving cells are sequenced. The cells are subjected to selective pressure for yet another period, after which a third sequencing procedure is performed. The cells can be subjected to the same selective pressure, followed by one or more further sequencing steps. Cells may be subjected to different selective pressures after the first (with or without additional sequencing steps). For example, surviving cells from the first selective pressure can be subjected to a second selective pressure to determine whether the surviving cells are sensitive or resistant to the second selective pressure. Cells may be subjected to multiple subjective pressures simultaneously, for example, to test a co-treatment.

[0059] Next, data from the sequencing process can be analyzed to determine which cellular traits make them more likely to survive, adapt, and / or be susceptible to selective pressures (one or more). For example, cells with increased (or decreased) expression of a particular RNA (or protein, or the presence of gene mutations / allelic / epigenetic profiles, etc.) may be more likely to survive and / or grow in the presence of selective pressures. The barcode corresponding to such cells may be expected to be present in higher abundances in surviving cells after exposure to selective pressures than the barcode of cells that do not have that level of expression (or presence). Conversely, barcodes corresponding to cells containing traits that make them less likely to survive and / or grow in the presence of selective pressures may be expected to be present in lower abundances (or absent) in surviving cells after exposure to selective pressures.

[0060] In one embodiment, a method is provided for screening cells for a trait. This method may include: (a) obtaining a plurality of barcoded cells, each containing a single unique barcode; (b) performing a first sequencing of RNA and / or DNA on a subset of the plurality of barcoded cells; (c) culturing the plurality of barcoded cells for a first period in the presence of selective pressure to form a first plurality of cells; (d) performing a second sequencing of RNA and / or DNA on a subset of the first plurality of cells; (e) culturing the first plurality of cells for a second period in the presence of selective pressure to form a second plurality of cells; (f) performing a third sequencing of RNA and / or DNA on at least a subset of the second plurality of cells; and (g) determining the level of the barcodes sequenced by the first sequencing, the second sequencing and / or the third sequencing. In embodiments, a single unique barcode is a unique combination of barcodes.

[0061] In another embodiment, a method is provided for screening cells for response traits to a therapeutic agent. This method may include: (a) obtaining a plurality of barcoded cells, each containing a single unique barcode; (b) performing a first sequencing of RNA and / or DNA on a subset of the plurality of barcoded cells; (c) culturing the plurality of barcoded cells for a first period in the presence of the therapeutic agent to form a first plurality of cells; (d) performing a second sequencing of RNA and / or DNA on a subset of the first plurality of cells; (e) culturing the first plurality of cells with the therapeutic agent for a second period to form a second plurality of cells; (f) performing a third sequencing of RNA and / or DNA on a subset of the second plurality of cells; and (g) determining the level of the barcodes sequenced in the first sequencing, the second sequencing and / or the third sequencing. In embodiments, a single unique barcode is a unique combination of barcodes.

[0062] With respect to the methods provided herein, in embodiments, the multiple barcoded cells of step (a) are grown in culture before step (b) or step (c). In embodiments, the multiple barcoded cells are grown in culture before step (a).

[0063] In embodiments, steps (e) and (f) are repeated one or more times, thereby forming one or more subsequent cells, and one or more subsequent sequencing is performed on a subset(s) of the subsequent cells. In embodiments, step (g) further includes determining the level of the sequenced barcode in one or more subsequent sequencing.

[0064] In an embodiment, the multiple barcoded cells comprise multiple clonal populations, where each cell within a single clonal population comprises the same single unique barcode. In an embodiment, the relative abundance of cells in each clonal population is approximately equal to the number of cells in each other clonal population in step (a). In an embodiment, the relative abundance of cells in each clonal population is determined relative to the number of cells containing each barcode determined in the first sequencing step, according to claim 7. In an embodiment, the single unique barcode is a unique combination of barcodes.

[0065] With respect to the methods provided herein, the levels of two or more barcodes sequenced by a first sequencing, a second sequencing, and / or a third sequencing can be determined. In embodiments, the levels of two or more barcodes sequenced by a first sequencing, a second sequencing, and / or a third sequencing are determined. In embodiments, the levels of two or more barcodes sequenced by a first sequencing and a second sequencing are determined. In embodiments, the levels of two or more barcodes sequenced by a second sequencing and a third sequencing are determined. In embodiments, the levels of two or more barcodes sequenced by a first sequencing and a third sequencing are determined. The barcode levels can be any determination of barcode quantity, such as barcode abundance, relative abundance, etc.

[0066] In embodiments, the methods provided herein further include identifying barcodes(s) that are concentrated in a first plurality of cells and / or a second plurality of cells and / or one or more subsequent plurality of cells. In embodiments, barcodes(s) are concentrated in a first plurality of cells and / or a second plurality of cells and / or one or more subsequent plurality of cells compared to one or more other barcodes. In embodiments, barcodes(s) are concentrated in comparison to the amount of barcodes(s) in a plurality of barcoded cells.

[0067] In embodiments, the methods provided herein further include identifying one or more genes that have higher levels of expression (and / or presence of traits such as gene mutations / allelation / epigenetic profiles) in cells containing enriched barcodes(one or more). In embodiments, identifying one or more genes (or presence of traits) includes determining that the expression level of at least one gene is higher (or the trait is present or more likely to be present) in cells containing enriched barcodes(one or more) compared to cells containing different barcodes(one or more) enriched or not enriched at different levels in a first sequencing, a second sequencing, a third sequencing, and / or one or more subsequent sequencing. In embodiments, identification includes identifying adaptive traits based on the higher expression (or increased presence of traits) of at least one gene. For example, cellular pathways may be associated by upregulation and / or downregulation of multiple genes involved in the pathway. In embodiments, identification involves identifying that a gene (trait) is involved in the cell's adaptation to a selective pressure or therapeutic agent based on higher expression of at least one gene (or increased presence of the trait).

[0068] In embodiments, the methods provided herein further include identifying one or more genes that have lower levels of expression (and / or absence of traits such as gene mutations / allelation / epigenetic profiles) in cells containing enriched barcodes (one or more) compared to cells containing different barcodes (one or more) with different levels of enrichment. In embodiments, identifying one or more genes with lower levels of expression (or absence of traits) includes determining that the expression level of at least one gene is lower (or the trait is absent or less likely to be present) in cells containing enriched barcodes (one or more) compared to cells containing different barcodes (one or more) enriched or not enriched at different levels in a first sequencing, a second sequencing, a third sequencing, and / or one or more subsequent sequencing. In embodiments, identification includes identifying adaptive traits based on the lower expression (or absence of traits) of at least one gene. In embodiments, determining involves identifying that a gene (trait) is involved in the cell's adaptation to a selective pressure or therapeutic agent based on the lower expression of at least one gene (or the absence of the trait).

[0069] In embodiments, the methods provided herein further include: (h) identifying barcodes(one or more) enriched in a first plurality of cells and / or a second plurality of cells and / or one or more subsequent plurality of cells; (i) identifying, from a first sequencing, a first gene having a higher expression level in cells containing enriched barcodes(one or more) than in cells containing barcodes(one or more) that are not enriched in the first plurality of cells and / or a second plurality of cells and / or one or more subsequent plurality of cells, wherein the higher expression of the first gene indicates that the first gene is an existing trait of candidate for resistance or sensitivity to selective pressure or therapeutic agents. In embodiments, the presence or absence of a trait is determined not by gene expression.

[0070] In embodiments, the methods provided herein further include: (j) identifying barcodes(one or more) enriched in a first plurality of cells and / or a second plurality of cells and / or one or more subsequent plurality of cells; (k) identifying from a first sequencing a second gene having a lower expression level in cells containing enriched barcodes(one or more) than in cells containing barcodes(one or more) that are not enriched in the first plurality of cells and / or a second plurality of cells and / or one or more subsequent plurality of cells, the lower expression of the second gene indicating that the second gene is an existing trait of candidate for resistance or sensitivity to selective pressure or therapeutic agents. In embodiments, the presence or absence of a trait is determined not by gene expression.

[0071] In embodiments, the methods provided herein further include: (l) identifying barcodes enriched in a first plurality of cells and / or a second plurality of cells and / or one or more subsequent plurality of cells; (m) identifying, from a second sequencing, a third gene having a higher expression level in cells containing enriched barcodes(one or more) than in cells containing barcodes that are not enriched in the first plurality of cells and / or a second plurality of cells and / or one or more subsequent plurality of cells, and the higher expression of the third gene indicates that the third gene is a candidate adaptive trait for resistance or sensitivity to selective pressure or therapeutic agents. In embodiments, the expression of the third gene is not high in cells containing barcodes(one or more) that are not enriched in the first plurality of cells and / or a second plurality of cells and / or one or more subsequent plurality of cells. In embodiments, the presence or absence of a trait is determined by gene expression rather than by the expression of the trait.

[0072] In embodiments, the methods provided herein further include: (n) identifying barcodes (one or more) enriched in a first plurality of cells and / or a second plurality of cells and / or one or more subsequent plurality of cells; (o) identifying, from a second sequencing and / or a third sequencing, a fourth gene having a lower expression level in cells containing enriched barcodes (one or more) than in cells containing barcodes (one or more) that are not enriched in the first plurality of cells and / or a second plurality of cells and / or one or more subsequent plurality of cells, wherein the lower expression of the fourth gene indicates that the first gene is a candidate adaptive trait for resistance or sensitivity to selective pressure or therapeutic agents. In embodiments, the expression of the fourth gene is not low in cells containing barcodes (one or more) that are not enriched in the first plurality of cells and / or a second plurality of cells. In embodiments, the presence or absence of a trait is determined by gene expression rather than by the expression of the trait.

[0073] In embodiments, with respect to the methods provided herein, one or more genes comprise a gene set. In embodiments, the gene set may be a gene set for tyrosine kinase inhibitor resistance, gap junctions, pathological Escherichia coli infection, hepatocellular carcinoma, DNA replication, carbon metabolism, cell cycle, glyoxylate and dicarboxylate metabolism, estrogen signaling pathways, fluid shear stress and atherosclerosis, histidine metabolism, Epstein-Barr virus infection, protein processing in the ER, metabolic pathways, lysosomes, focal adhesion, ECM-receptor interactions, small cell lung cancer, apoptosis and / or integrated stress response.

[0074] In embodiments, with respect to the methods provided herein, the unique barcode may or may not be incorporated into the genome of the barcoding cell. In embodiments, the unique barcode is incorporated into the genome of the barcoding cell. In embodiments, the unique barcode is not incorporated into the genome of the barcoding cell. In embodiments, the barcode is expressed by the cell. For example, the barcode may be adjacent to the coding sequence so as to be expressed together with the coding sequence. The barcode may be expressed as RNA. In embodiments, the amount of barcode in a cell population is determined based on the abundance of the barcode sequence determined by NGS. In embodiments, the barcode is not expressed by the cell. For example, the amount of barcode in a cell population may be determined based on DNA sequencing or other DNA analysis (e.g., PCR) for the presence / absence / amount of barcode DNA in the cell population. For example, the amount of barcode in a population may be determined based on DNA analysis other than DNA sequencing or conduction DNA sequencing. In embodiments, the unique barcode is a unique combination of barcodes.

[0075] With respect to the methods provided herein, the first sequencing, second sequencing, third sequencing, and / or one or more subsequent sequencing may be performed using RNA-seq, DNA sequencing, epigenetic sequencing, or protein sequencing. In embodiments, the first sequencing, second sequencing, third sequencing, and / or one or more subsequent sequencing may be performed using next-generation sequencing (NGS). In embodiments, the first sequencing, second sequencing, third sequencing, and / or one or more subsequent sequencing may be performed using RNA-seq. In embodiments, RNA-seq is single-cell RNA-seq. In embodiments, the first sequencing, second sequencing, third sequencing, and / or one or more subsequent sequencing may be DNA sequencing. In embodiments, the first sequencing, second sequencing, third sequencing, and / or one or more subsequent sequencing are epigenetic sequencing. In embodiments, the first sequencing, second sequencing, third sequencing, and / or one or more subsequent sequencing are protein sequencing. In embodiments, with respect to the methods provided herein, the first sequencing, second sequencing, third sequencing, and / or one or more subsequent sequencing may exclude any one of the sequencing methods enumerated herein. In embodiments, the first sequencing, second sequencing, third sequencing, and / or one or more subsequent sequencing are not performed using one or more of RNA-seq, DNA sequencing, epigenetic sequencing, or protein sequencing. In the embodiment, the first sequencing, the second sequencing, the third sequencing, and / or one or more subsequent sequencing are not RNA-seq.In the embodiment, the first sequence, the second sequence, the third sequence, and / or one or more subsequent sequences are not DNA sequences. In the embodiment, the first sequence, the second sequence, the third sequence, and / or one or more subsequent sequences are not epigenetic sequences. In the embodiment, the first sequence, the second sequence, the third sequence, and / or one or more subsequent sequences are not protein sequences. In the embodiment, the first sequence, the second sequence, the third sequence, and / or one or more subsequent sequences are not conduction DNA sequences. In the embodiment, the first sequence is not conduction DNA sequence. In the embodiment, the second sequence is not conduction DNA sequence. In the embodiment, the third sequence is not conduction DNA sequence. In the embodiment, one or more subsequent sequencing steps are not conduction DNA sequencing.

[0076] In embodiments, with respect to the methods provided herein, the selective pressure is treatment with a therapeutic agent, contact with a contaminant, genomic engineering, engraftment into a host, culture conditions, growth conditions, contact with a stimulus, or contact with other cells. In embodiments, the selective pressure is treatment with a therapeutic agent. In embodiments, the selective pressure is contact with a contaminant. In embodiments, the selective pressure is genomic engineering. In embodiments, the selective pressure is engraftment into a host. In embodiments, the selective pressure is culture conditions. In embodiments, the selective pressure is growth conditions. In embodiments, the selective pressure is contact with a stimulus. In embodiments, the selective pressure is contact with other cells. In embodiments, one or more selective pressures may be excluded.

[0077] In embodiments, with respect to the methods provided herein, the first period is about 30 minutes to about 1 month. In embodiments, the first period is about 30 minutes. In embodiments, the first period is about 1 hour. In embodiments, the first period is about 5 hours. In embodiments, the first period is about 10 hours. In embodiments, the first period is about 15 hours. In embodiments, the first period is about 20 hours. In embodiments, the first period is about 1 day. In embodiments, the first period is about 2 days. In embodiments, the first period is about 3 days. In embodiments, the first period is about 4 days. In embodiments, the first period is about 5 days. In embodiments, the first period is about 6 days. In embodiments, the first period is about 7 days. In embodiments, the first period is about 8 days. In embodiments, the first period is about 9 days. In one embodiment, the first period is approximately 10 days. In another embodiment, the first period is approximately 11 days. In another embodiment, the first period is approximately 12 days. In another embodiment, the first period is approximately 13 days. In another embodiment, the first period is approximately 14 days. In another embodiment, the first period is approximately 15 days. In another embodiment, the first period is approximately 16 days. In another embodiment, the first period is approximately 17 days. In another embodiment, the first period is approximately 18 days. In another embodiment, the first period is approximately 19 days. In another embodiment, the first period is approximately 20 days. In another embodiment, the first period is approximately 21 days. In another embodiment, the first period is approximately 22 days. In another embodiment, the first period is approximately 23 days. In another embodiment, the first period is approximately 24 days. In another embodiment, the first period is approximately 25 days. In one embodiment, the first period is approximately 26 days. In another embodiment, the first period is approximately 27 days. In yet another embodiment, the first period is approximately 28 days. In yet another embodiment, the first period is approximately 29 days. In yet another embodiment, the first period is approximately 30 days. In yet another embodiment, the first period is approximately 1 month. In yet another embodiment, the first period is longer than 1 month.In embodiments, the first period is about 30 minutes, about 1 hour, about 5 hours, about 10 hours, about 15 hours, about 20 hours, about 1 day, about 2 days, about 3 days, about 4 days, about 5 days, about 6 days, about 7 days, about 8 days, about 9 days, about 10 days, about 11 days, about 12 days, about 13 days, about 14 days, about 15 days, about 16 days, about 17 days, about 18 days, about 19 days, about 20 days, about 21 days, about 22 days, about 23 days, about 24 days, about 25 days, about 26 days, about 27 days, about 28 days, about 29 days, about 30 days, or about 1 month. The amount of time may be any value or a partial range within the range provided herein and includes the endpoint.

[0078] In embodiments, with respect to the methods provided herein, the second period is about 12 hours to about 12 months. In embodiments, the second period is about 12 hours. In embodiments, the second period is about 1 day. In embodiments, the second period is about 5 days. In embodiments, the second period is about 10 days. In embodiments, the second period is about 15 days. In embodiments, the second period is about 20 days. In embodiments, the second period is about 25 days. In embodiments, the second period is about 1 month. In embodiments, the second period is about 1.5 months. In embodiments, the second period is about 2 months. In embodiments, the second period is about 2.5 months. In embodiments, the second period is about 3 months. In embodiments, the second period is about 3.5 months. In embodiments, the second period is about 4 months. In an embodiment, the second period is approximately 4.5 months. In an embodiment, the second period is approximately 5 months. In an embodiment, the second period is approximately 5.5 months. In an embodiment, the second period is approximately 6 months. In an embodiment, the second period is approximately 6.5 months. In an embodiment, the second period is approximately 7 months. In an embodiment, the second period is approximately 7.5 months. In an embodiment, the second period is approximately 8 months. In an embodiment, the second period is approximately 8.5 months. In an embodiment, the second period is approximately 9 months. In an embodiment, the second period is approximately 9.5 months. In an embodiment, the second period is approximately 10 months. In an embodiment, the second period is approximately 10.5 months. In an embodiment, the second period is approximately 11 months. In an embodiment, the second period is approximately 11.5 months. In an embodiment, the second period is approximately 12 months. In an embodiment, the second period is approximately 12 hours, approximately 1 day, approximately 5 days, approximately 10 days, approximately 15 days, approximately 20 days, approximately 25 days, approximately 1 month, approximately 1.5 months, approximately 2 months, approximately 2.5 months, approximately 3 months, approximately 3.5 months, approximately 4 months, approximately 4.5 months, approximately 5 months, approximately 5.5 months, approximately 6 months, approximately 6.5 months, approximately 7 months, approximately 7.5 months, approximately 8 months, approximately 8.5 months, approximately 9 months, approximately 9.5 months, approximately 10 months, approximately 10.5 months, approximately 11 months, approximately 11.5 months, or approximately 12 months.In the embodiment, the second period is longer than one year. The time amount may be any value or partial range within the range provided herein and includes the endpoint.

[0079] In one embodiment, a method is provided for comparing responses to selective pressure. This method includes: (a) obtaining a first plurality of barcoded cells, each barcoded cell containing a single unique barcode; (b) obtaining a second plurality of barcoded cells substantially similar to the first plurality of barcoded cells; (c) performing a first sequencing of RNA and / or DNA from the first plurality of barcoded cells and / or the second plurality of barcoded cells; (d) culturing the first plurality of barcoded cells in the presence of a first selective pressure to form the first plurality of cells; (e) culturing the second plurality of barcoded cells in the presence of a second selective pressure to form the second plurality of cells; (f) (g) performing a second sequencing of RNA and / or DNA from a first plurality of cells and / or a second plurality of cells; (h) culturing the first plurality of cells in the presence of a first selective pressure to form a third plurality of cells; (i) culturing the second plurality of cells in the presence of a second selective pressure to form a fourth plurality of cells; (j) performing a third sequencing of RNA and / or DNA from the third plurality of cells and / or a fourth plurality of cells; and (j) determining the levels of one or more barcodes sequenced in the first sequencing, second sequencing and / or third sequencing. In embodiments, steps (g) to (i) are repeated in one or more iterations to form one or more subsequent plurality of cells and one or more subsequent sequencing. In embodiments, step (j) further includes determining the levels of one or more barcodes in one or more subsequent sequencing steps. In embodiments, a single unique barcode is a unique combination of barcodes. In embodiments, with respect to the methods provided herein, the first sequencing, second sequencing, third sequencing, and / or one or more subsequent sequencing may exclude any one of the sequencing methods enumerated herein.In the embodiments, the first sequencing, second sequencing, third sequencing, and / or one or more subsequent sequencing are not performed using one or more of RNA-seq, DNA sequencing, epigenetic sequencing, or protein sequencing. In the embodiments, the first sequencing, second sequencing, third sequencing, and / or one or more subsequent sequencing are not RNA-seq. In the embodiments, the first sequencing, second sequencing, third sequencing, and / or one or more subsequent sequencing are not DNA sequencing. In the embodiments, the first sequencing, second sequencing, third sequencing, and / or one or more subsequent sequencing are not epigenetic sequencing. In the embodiment, the first sequence, the second sequence, the third sequence, and / or one or more subsequent sequences are not protein sequences. In the embodiment, the first sequence, the second sequence, the third sequence, and one or more subsequent sequences are not conduction DNA sequences. In the embodiment, the first sequence is not conduction DNA sequences. In the embodiment, the second sequence is not conduction DNA sequences. In the embodiment, the third sequence is not conduction DNA sequences. In the embodiment, one or more subsequent sequences are not conduction DNA sequences.

[0080] In embodiments, with respect to the method provided herein, the first plurality of barcoded cells and the second plurality of barcoded cells comprise a plurality of clonal populations, where each cell in a single clonal population contains the same single unique barcode. In embodiments, the relative abundance of cells in each clonal population is approximately equal to the number of cells in each of the other clonal populations in steps (a) and (b). In embodiments, the relative abundance of cells in a clonal population is determined from the first sequencing step relative to the number of cells containing one or more barcodes. In embodiments, a single unique barcode is a unique combination of barcodes.

[0081] In embodiments, the method provided herein further includes determining a first expression level of a gene in cells having one or more barcodes enriched in a first plurality of cells and / or a third plurality of cells and / or a plurality of subsequent cells, based on a second sequencing and / or a third sequencing and / or one or more subsequent sequencing, and determining a second expression level of a gene in a second plurality of barcoded cells and / or a fourth plurality of cells and / or a plurality of subsequent cells, based on a first sequencing, a second sequencing and / or a third sequencing and / or one or more subsequent sequencing. In embodiments, the method further includes comparing the first expression level of a gene with the second expression level of a gene. In embodiments, the presence or absence of a trait is determined by the gene expression rather than by the gene expression itself.

[0082] In one embodiment, a method is provided for screening cells for phenotypes in the cells. This method comprises: (a) providing a mixture of cells comprising a plurality of clonal populations, wherein each clonal population comprises an identifier unique to that clonal population, and the initial genetic, transcriptome, and / or proteome information of at least one representative member of each clonal population is known; (b) culturing the mixture of cells in the presence of a first selective pressure for a first period, and at the end of the first period, obtaining second genetic, transcriptome, and / or proteome information of at least one member of a viable clonal population from the mixture of cells; (c) subjecting the mixture of cells subjected to the first selective pressure to a second selective pressure for a second period, and at the end of the second period, obtaining third genetic, transcriptome, and / or proteome information of at least one member of a viable clonal population from the mixture of cells; and (d) determining the level of a clonal population present in a final mixture of cells based on a unique identifier relating to the clonal population. In the embodiment, step (b) is repeated in one or more iterations. In the embodiment, step (c) is repeated in one or more iterations. In the embodiment, steps (b) and (c) are repeated in one or more iterations.

[0083] In embodiments, the methods provided herein further include identifying adaptive traits, where adaptive traits are genetic and / or proteomic traits present or absent in the clonal population in the final mixture of cells. In embodiments, adaptive traits are the presence or absence of a gene, allele, gene modification, transcription, or protein; or a change in a gene, allele, transcription, or protein when comparing the obtained first, second, or third, or one or more subsequent genetic, transcriptome, and / or proteomic information. In embodiments, adaptive traits are the presence of a gene, allele, gene modification, transcription, or protein. In embodiments, adaptive traits are the absence of a gene, allele, gene modification, transcription, or protein. In embodiments, adaptive traits are a change in a gene, allele, transcription, or protein when comparing the obtained first, second, or third, or one or more subsequent genetic, transcriptome, and / or proteomic information.

[0084] In an embodiment, step (b) includes obtaining a fourth genetic information, transcriptome information and / or proteome information of at least one member of a second surviving clonal population from a mixture of cells. In an embodiment, step (c) includes obtaining a fifth genetic information, transcriptome information and / or proteome information of at least one member of a second surviving clonal population from a mixture of cells.

[0085] In embodiments, the method further includes comparing the initial genetic information, transcriptome information, and / or proteome information, the second genetic information, transcriptome information, and / or proteome information, and / or the third genetic information, transcriptome information, and / or proteome information, and / or the information from one or more subsequent genetic information, transcriptome information, and / or proteome information. In embodiments, the method further includes comparing the initial genetic information, transcriptome information, and / or proteome information, the second genetic information, transcriptome information, and / or proteome information, and / or the third genetic information, transcriptome information, and / or proteome information, and / or the information from one or more subsequent genetic information, transcriptome information, and / or proteome information to the genetic information, transcriptome information, and / or proteome information of different clonal populations of cells having different unique barcodes that have been subjected to selective pressure. In embodiments, the method further includes comparing initial genetic, transcriptome, and / or proteome information, second genetic, transcriptome, and / or proteome information, and / or third genetic, transcriptome, and / or proteome information, and / or one or more subsequent genetic, transcriptome, and / or proteome information from a clonal population of cells having a unique barcode, with genetic, transcriptome, and / or proteome information from different determination steps relating to the clonal population of cells. In embodiments, the unique barcode is a unique combination of barcodes.

[0086] In embodiments, obtaining genetic information, transcriptome information and / or proteome information with respect to methods provided herein includes RNA-seq, DNA sequencing, epigenetic sequencing, or protein sequencing. In embodiments, obtaining genetic information, transcriptome information and / or proteome information includes NGS. In embodiments, obtaining genetic information, transcriptome information and / or proteome information includes RNA-seq. In embodiments, RNA-seq is single-cell RNA-seq. In embodiments, obtaining genetic information, transcriptome information and / or proteome information includes DNA sequencing. In embodiments, obtaining genetic information, transcriptome information and / or proteome information includes epigenetic sequencing. In embodiments, obtaining genetic information, transcriptome information and / or proteome information includes protein sequencing.

[0087] In embodiments, the first sequencing, second sequencing, third sequencing, and / or one or more subsequent sequencing in the methods provided herein may exclude any one of the sequencing methods enumerated herein. In embodiments, the first sequencing, second sequencing, third sequencing, and / or one or more subsequent sequencing are not performed using one or more of RNA-seq, DNA sequencing, epigenetic sequencing, or protein sequencing. In embodiments, the first sequencing, second sequencing, third sequencing, and / or one or more subsequent sequencing are not RNA-seq. In embodiments, the first sequencing, second sequencing, third sequencing, and / or one or more subsequent sequencing are not DNA sequencing. In the embodiments, the first sequence, the second sequence, the third sequence, and / or one or more subsequent sequences are not epigenetic sequences. In the embodiments, the first sequence, the second sequence, the third sequence, and / or one or more subsequent sequences are not protein sequences. In the embodiments, the first sequence, the second sequence, the third sequence, and one or more subsequent sequences are not conduction DNA sequences. In the embodiments, the first sequence is not conduction DNA sequences. In the embodiments, the second sequence is not conduction DNA sequences. In the embodiments, the third sequence is not conduction DNA sequences. In the embodiments, one or more subsequent sequences are not conduction DNA sequences.

[0088] In embodiments relating to the methods provided herein, the first selective pressure and / or the second selective pressure include treatment with a therapeutic agent, contact with a contaminant, genomic engineering, engraftment into a host, culture conditions, growth conditions, contact with a stimulus, or contact with other cells.

[0089] In one embodiment, a method is provided for identifying cellular programs that facilitate adaptation to pressure. This method includes: (a) transducing cells with multiple barcodes such that each cell contains a single unique barcode; (b) growing the cells in culture to create a starting cell pool of clones of cells containing each barcode; (c) obtaining first genetic information, transcriptome information, and / or proteome information from a first subset of the starting cell pool; (d) growing a second subset of the starting cell pool in the presence of selective pressure to form an intermediate cell pool; (e) obtaining second genetic information, transcriptome information, and / or proteome information from a first subset of the intermediate cell pool; (f) continuing to culture a second subset of the intermediate cell pool in the presence of selective pressure to form an intermediate (g) growing a cell pool to form a final cell pool; (h) obtaining third genetic information, transcriptome information, and / or proteome information from at least a subset of the final cell pool; (i) assigning cells with enriched barcodes in the final cell pool and / or starting cell pool as winning clones and / or assigning cells with depleted barcodes in the final cell pool as losing clones; and (j) determining gene mutations, transcriptional programs, and / or protein expression (and / or other traits) associated with at least one winning clone and / or at least one losing clone. In embodiments, a single unique barcode is a unique combination of barcodes. In embodiments, a starting cell pool is created using approximately equal numbers of clones of cells containing each barcode. In embodiments, the relative abundance of cell clones is normalized to the number of cells containing each barcode obtained in step (c). In the embodiment, steps (d) and (e) are repeated one or more times to obtain one or more additional intermediate cell pools and one or more additional intermediate genetic information, transcriptome information, and / or proteome information.In an embodiment, step (h) further includes quantifying the level of each barcode in one or more additional intermediate cell pools.

[0090] With respect to the methods provided herein, in embodiments, selective pressure includes treatment with a therapeutic agent, contact with a contaminant, genomic engineering, engraftment into a host, culture conditions, growth conditions, contact with a stimulus, or contact with other cells. In embodiments, selective pressure includes treatment with a therapeutic agent. In embodiments, selective pressure includes contact with a contaminant. In embodiments, selective pressure includes genomic engineering. In embodiments, selective pressure includes engraftment into a host. In embodiments, selective pressure includes culture conditions. In embodiments, selective pressure includes growth conditions. In embodiments, selective pressure includes contact with a stimulus. In embodiments, selective pressure includes contact with other cells.

[0091] As described above, the methods provided herein may include obtaining genetic information, transcriptome information, and / or proteome information. In embodiments, obtaining genetic information, transcriptome information, and / or proteome information includes NGS. In embodiments, obtaining genetic information, transcriptome information, and / or proteome information includes RNA-seq. In embodiments, RNA-seq is single-cell RNA-seq. In embodiments, obtaining genetic information, transcriptome information, and / or proteome information includes DNA sequencing. In embodiments, obtaining genetic information, transcriptome information, and / or proteome information includes epigenetic sequencing. In embodiments, obtaining genetic information, transcriptome information, and / or proteome information includes protein sequencing. In embodiments, with respect to the methods provided herein, obtaining genetic information, transcriptome information, and / or proteome information may exclude any one of the sequencing methods enumerated herein. In embodiments, obtaining genetic information, transcriptome information, and / or proteome information does not involve NGS. In embodiments, obtaining genetic information, transcriptome information, and / or proteome information does not involve RNA-seq. In embodiments, obtaining genetic information, transcriptome information, and / or proteome information does not involve DNA sequencing. In embodiments, obtaining genetic information, transcriptome information, and / or proteome information does not involve epigenetic sequencing. In embodiments, obtaining genetic information, transcriptome information, and / or proteome information does not involve protein sequencing. [Examples]

[0092] Examples Those skilled in the art will understand that the descriptions of the preparation and use of particles provided herein are for illustrative purposes only and that this disclosure is not limited by such examples.

[0093] Example 1. Exemplary workflow for longitudinal scRNA-seq transcriptome analysis Figure 2 shows an exemplary workflow. The lentiviral vector, for example as shown in Figure 3, contains a barcode insertion site in the 3'UTR of the Puro-IRES-GFP transgene and can be used to create a barcoded vector library with a desired number (e.g., 1 million) different barcodes. Lentiviral particles can be produced using a standard virus production protocol, and the target cell population can be transduced with an infection multiplicity (MOI) of 0.1 or less. This MOI has been previously optimized to yield a single barcode per cell. Generally, infection of 20 to 50 million cells yields a total of 2 to 5 million infected cells. Infected cells can be double-selected by culturing the cells in the presence of puromycin followed by fluorescence-activated cell sorting (FACS) for EGFP.

[0094] To establish a barcoded pool with clones labeled by unique barcodes, transduced cells can be dissociated into single cells and counted. N cells (less than 1% of the infected cells in the second step) can be seeded into cell culture dishes at approximately 1 cell per well and grown to the desired number to obtain multiple cells consisting of N single-cell clones. For example, the starting cell pool may contain approximately N uniquely barcoded single-cell clones. More than 20N cells in the starting pool can be profiled, for example, by single-cell RNA sequencing (scRNA-seq, e.g., using the 10xTechnology platform).

[0095] Simultaneously, more than 20N cells from the starting pool can be subjected to the desired selection pressure, preferably through at least three technical replications for each condition. The pool of more than 20N cells can be grown under the desired selection and profiled at further time points by scRNA-seq. After selection is complete, the cells that survive to the end can be profiled again by scRNA-seq. The surviving cells can also be subjected to genomic DNA sequencing, such as final barcode count quantification by next-generation sequencing (NGS) analysis. Cells with barcodes enriched at the endpoint can be assigned as "winner" clones. Cells with barcodes depleted at the endpoint can be assigned as "loser" clones. Cells with barcodes that proliferate the most during selection can be assigned as "superadapters".

[0096] The transcriptional programs associated with winning clones, losing clones, and hyperadaptive clones can be determined by aggregating scRNA-seq data obtained before, during, and after selection. Transcriptional features associated with the winning clone before treatment can be identified as candidate existing programs that promote survival and adaptation with respect to the selected clone. Transcriptional features associated with the hyperadaptive clone during treatment can be identified as candidate adaptive programs that promote adaptation.

[0097] Furthermore, the same starting pool can be subjected to different treatments to compare the selective pressures exerted by these treatments, and / or the responses and adaptive processes induced by these different treatments.

[0098] Example 2: Identification of existing non-genetic characteristics of EGFRi-resistant cells Using the workflow described in Example 1, the pre-existing non-genetic features of EGFR inhibitor-resistant PC9 (non-small cell lung carcinoma) cells were identified without the need to purify the resistant population. The study design is shown in Figure 6. 600 PC9 cells, each containing a unique barcode, were grown separately for 12 days. Clones were then pooled into a cell population with replicated barcodes and subjected to a first scRNA-seq on day 1. Subsequently, the cell population was treated with either DMSO (vehicle control) or erlotinib. On day 5 of treatment, a second scRNA-seq was performed to evaluate the initial response of the cells to the treatment. As shown in Figure 7, the initial response of PC9 cells to erlotinib was robust, with a survival rate reduction of over 98% within 5 days compared to the DMSO control. Treatment with erlotinib was continued for approximately two more months. At the treatment endpoint, a final scRNA-seq was performed on viable cells, yielding 4 × 10⁶ nucleotides. 6 The number exceeded that of a single cell.

[0099] The pre-existing traits of the barcoded cells enriched on day 5 were evaluated by analyzing day 1 sequencing data of the relevant barcoded cells shown in Figure 8. Existing EGFR inhibitor-resistant populations are known in the art and include a genetically pre-existing T790M mutation occurring in approximately 1 in 50,000 cells and a genetically non-transferable population occurring in approximately 1 in 50 cells. Under these experimental conditions, these same non-genetic features and other genes upregulated in resistant clones before treatment were identified by transcriptome data analysis from day 1 scRNA-seq, although the much rarer genetic T790M mutation was not identified. These data demonstrate that pre-existing features conferring either susceptibility or resistance to different drug treatments can be rapidly identified using the methods described herein.

[0100] Example 3: Indication of lung cancer cells with erlotinib or degrader G-104 Using the workflow described above, we evaluated the adaptability of mutant EGFR lung cancer cells to treatment strategies including EGFR kinase inhibition (erlotinib) and EGFR proteolysis (Degrader G-104 or Degrader) using a PC9 (non-small cell lung cancer) cell line model. A representative experimental design is shown in Figure 6. 600 PC9 cells, each containing a unique barcode, were grown separately for 12 days. Clones were then pooled into a cell population with replicated barcodes and subjected to a first scRNA-seq on day 1. Subsequently, the cell population was treated with either DMSO (vehicle control), erlotinib, or Degrader. A second scRNA-seq was performed on day 5 of treatment to assess the initial response of the cells to the treatment. Treatment with erlotinib or Degrader was continued for approximately two more months. At the treatment endpoint, a final scRNA-seq was performed on viable cells. Using next-generation sequencing technology, scRNA-seq data were analyzed, and barcode quantities were determined for both treatments, as shown in Figures 4A-4B. Analysis of barcode quantification showed that a subset of cell clones exhibited a difference in viability when subjected to erlotinib treatment compared to degrader treatment. As shown in Figure 14, the barcode subset was preferentially depleted by erlotinib compared to degrader treatment.

[0101] The pre-existing characteristics of barcoded cells enriched on day 5 were evaluated by analyzing day 1 sequencing data of the relevant barcoded cells. Day 1 transcriptome data showed higher expression levels of genes involved in epithelial-mesenchymal transition (EMT) and EGFRi resistance signatures in barcoded cells enriched in both the erlotinib-treated and degrader-treated groups. Surviving cells in the degrader-treated group had pre-existing low expression levels of proteasome genes and endoplasmic reticulum (ER) protein processing genes. These results indicate that cells with higher levels of expression of genes related to migratory and invasive characteristics confer insensitivity to treatment with both erlotinib and degrader. Furthermore, the results are consistent with existing knowledge in the field that treatment with degrader induces receptor internalization, and cells with low expression levels of proteasome and ER protein processing genes can mobilize enzymes for proteolysis to reduce ER stress, thereby avoiding cell death and adapting to degrader.

[0102] Heatmap analysis of transcriptome data at day 5 from clones with enriched barcodes at day 5 shows that treatment with erlotinib induced higher expression levels of genes involved in endoplasmic reticulum stress response and ER protein processing in cells adapted at day 5 compared to cells treated with a degrader. This is consistent with knowledge in the art that a subset of cancer cells restore protein processing and mitigate ER stress as a survival mechanism. In contrast, higher expression levels of the same combination of genes sensitized cells to the degrader, resulting in a decrease in barcodes associated with high levels of expression.

[0103] Studies using lung cancer cell lines treated with degraders and kinase inhibitors showed that degraders promote resistance to EGFR inhibition and further attenuate kinase inhibitor activity. H1975 cells were first treated with 100 nM osimertinib, a kinase inhibitor, for 4 hours, and then co-treated with 100 nM osimertinib + 1 μM degrader for 24 hours. As shown in Figure 12, the addition of degraders resulted in an attenuation of kinase inhibitor activity in H1975 cells. These results indicate that the MOA of conventional kinase inhibitors and degraders varies in mediating therapeutic efficacy and confirm the different survivals of clonal populations treated with degraders versus erlotinib.

[0104] Degrader-resistant and susceptible clonal populations were identified using an experimental setup similar to that shown in Figure 6. Figure 9 shows a heatmap analysis of day 5 transcriptome data from clones with depleted or enriched barcodes. Analysis of single-cell RNA-seq transcriptome data allowed for the identification of specific pathways upregulated in degrader-resistant clones and confirmed that degrader resistance is driven by specific mechanisms distinct from those known to confer resistance to EGFR kinase inhibitors.

[0105] Further analysis of single-cell RNA-seq data was conducted to assess how degrader-resistant clones exhibited different responses to erlotinib versus degrader. Figure 10A shows a heatmap analysis of day 5 transcriptome data comparing erlotinib-affected pathways with degrader-affected pathways in degrader-resistant clones. Compared to degrader, erlotinib activated genes involved in the ER protein processing pathway. Expression of ATF4 and SLC7A5, genes associated with ER stress-mediated cell death and resulting in integrated stress response (ISR) induction, was further upregulated on day 5 of erlotinib treatment, as shown in Figure 10B. These results indicate that erlotinib can function as a cytotoxic agent by activating ER protein processing and ISR pathways. Furthermore, erlotinib was shown to downregulate pathways including DNA replication, carbon metabolism, and the cell cycle, which are consistent with and reflect the proliferation arrest induced by erlotinib.

[0106] Since the ISR pathway was shown to be activated by treatment with erlotinib rather than a degrader, we modeled proteolysis by investigating the depletion of EGFR protein production. Therefore, PC9 cells were treated with either siNTC + erlotinib, siEGFR + erlotinib, or siEGFR + DMSO, and then ISR and MAPK genes were analyzed (via scRNA-seq) after 72 hours, and cell viability was assessed on day 6. The experimental setup is shown in Figure 11A. The results shown in Figure 11B indicate that acute depletion of newly synthesized EGFR protein leads to downregulation of gene expression related to ER stress and ISR pathway induction. As shown in Figure 11C, both erlotinib and the degrader strongly repressed standard MAPK target genes, so MAPK gene expression was used as a control. The results shown in Figure 11D indicate that erlotinib was less potent when combined with siEGFR, as evidenced by the higher number of viable cells in the erlotinib + siEGFR treatment group compared to treatment with erlotinib alone. In summary, these results demonstrate that inhibition of EGFR protein production by degraders weakens the efficacy of ISR induction and EGFR inhibitors.

[0107] To further investigate the MOA revealed by scRNA-seq transcriptome data, pharmacological modulation of ISR was tested with degrader or erlotinib. Figure 13A shows a schematic diagram of proteins involved in ISR induction and pharmacological modulation. Tunicamycin and thapsigargin were used as inducers of ER stress, and integrated stress response inhibitors (ISRIBs) were used to inhibit ISR activation. Cells were treated with degrader, degrader + tunicamycin, or degrader + thapsigargin to evaluate the effect of ER stress on the efficacy of degrader. In a separate experimental group, cells were treated with either degrader, erlotinib, or erlotinib and ISRIB. The results shown in Figure 13B indicate that the combination of degrader and the induction of mild ER stress leads to enhanced efficacy of degrader. This confirms the results from scRNA-seq data analysis that degrader itself does not activate the ISR pathway. As shown in Figure 13C, the addition of ISRIB to erlotinib resulted in a loss of erlotinib efficacy, which is demonstrated by increased cell viability. Further confirmation of ISRIB-mediated inhibition of erlotinib ISR gene activation was achieved by qPCR analysis, as shown in Figure 15.

[0108] The experiments described above highlight the ability to track barcoded cells using the scRNA-seq platform and elucidate both drug MOA and resistance mechanisms. Using the methods described herein, it has been shown that EGFR kinase inhibitors function by both inhibiting EGFR and activating the ISR pathway, as shown in Figure 16, thereby exerting cytotoxic effects through a bifurcated pathway.

[0109] These data demonstrate that it is possible to track the time-course response of a clonal population to a treatment of interest and to correlate existing characteristics conferring susceptibility versus insensitivity, programs promoting or preventing adaptation, and the final adaptive state and outcome. This can be done at single-cell resolution. Simultaneously, this technique also utilizes high-throughput technology, thereby more accurately capturing transcriptional programs by aggregating scRNA-seq data collected from cells within the same clone and / or clones with similar adaptive trajectories. In contrast, previous models only compare the starting population with the selected population and therefore cannot resolve existing traits for adaptive change.

[0110] Example 4: Identification of resistance mechanisms in KRAS cancer and development of combination therapies While G12C mutations in KRAS are known to play a crucial role in aggressive cancers, therapeutic agents that inhibit mutant KRAS have proven challenging. The workflow described above is applied to elucidate resistance mechanisms to mutant KRAS inhibitors and to inform the development of protocols for administering KRAS inhibitors in combination with other therapeutic agents. Key KRASG12C mutant inhibitor resistance mechanisms in preclinical models include non-genetic features as well as endogenous and / or innate characteristics that determine KRAS dependence. Using the methods described above, we will address how resistant cells respond differently to KRAS inhibitors compared to susceptible cells in heterogeneous populations and help identify which genes and / or pathways enable early cell survival. Furthermore, this method will help identify the mechanisms that surviving cells use to adapt to KRAS inhibition and become completely resistant. In summary, these results will help determine which drug combinations are most effective in eliminating existing resistant cells.

[0111] Example 5: TraCe-Seq reveals existing adaptive characteristics underlying the unexpectedly inferior efficacy of targeted EGFR degradation compared to inhibition. Genetic and non-genetic heterogeneity within cancer cell populations poses a significant challenge to anti-cancer therapies. A robust paradigm is currently lacking to address how existing adaptive features influence cellular responses to treatment. Here, we developed the TraCe-seq method, which combines clonal tracking and single-cell RNA sequencing to capture, at clonal resolution, the origin, fate, and differential early adaptation programs of cells within a composite population in response to different treatments.

[0112] Using TraCe-seq, we benchmarked how next-generation dual EGFR inhibitor-degraders compare to standard EGFR kinase inhibitors in EGFR-mutated lung cancer cells. We identified a paradoxical loss of antiproliferative activity associated with targeted degradation of the EGFR protein, and an unexpected essential role of the ER protein processing pathway in anti-EGFR therapeutic efficacy. This exemplary study challenges the assumption that targeted degradation is universally superior to enzyme inhibition and demonstrates TraCe-seq as a broadly applicable approach for studying how existing transcriptional programs influence treatment responses.

[0113] Targeted therapies for oncogenic driver mutations offer significant clinical benefits to cancer patients and represent great potential for precision medicine. However, not all patients with such mutations respond equally. While resistance mechanisms such as secondary site mutations have been reported, other existing and acquired resistance-conferring mechanisms pose significant challenges to overall clinical response and durability, even among the best-studied therapies.

[0114] The treatment options for cancer are constantly evolving to enhance responses and overcome resistance. While direct inhibition of oncogenic drivers (perhaps best exemplified by the success of small molecule inhibitors of kinases) remains an integral part of precision therapy, new therapeutic modalities are also being studied. In recent years, there has been growing interest in novel mechanisms of action (MOAs) of targeted proteolysis rather than conventional occupation-based targeted inhibition. Heterobifunctional targeted protein degraders, molecules that can simultaneously recruit E3 ubiquitin ligases to a target of interest and induce targeted degradation via ubiquitin-mediated proteolysis, have attracted considerable attention and have been shown to be superior to enzyme inhibition alone in certain situations. However, it remains unclear whether dual-action inhibitors / degraders offer a universal advantage.

[0115] The current paradigm for characterizing drug response and resistance relies heavily on endpoint assessment after long-term drug exposure. While this process may be effective in identifying resistance mechanisms, it has limitations in its ability to reveal pre-existing features and early adaptive changes that enable the survival of specific subpopulations. A system is needed that simultaneously tracks origins and compares the immediate response / fate of tumor cells to different therapies, which could significantly accelerate drug response / resistance research. This would allow for direct comparisons of the efficacy and MOA of different therapies, thus further accelerating the development of future treatments. To this end, a method called TraCe-seq (Tracking Differential Clonal Response by scRNA-sequencing) was developed to rapidly capture the underlying origins and early adaptive processes of treatment responses by simultaneously measuring clonal compatibility and their transcriptional trajectories in heterogeneous populations treated with anti-cancer therapy (Figure 17A). As demonstrated below, this method enables direct and comprehensive comparisons of different treatment modalities at subpopulation and single-cell resolution, providing novel insights into pre-existing transcriptional features that determine drug response and subsequent resistance.

[0116] To set up TraCe-seq, a barcoding library based on a 3' scRNA-seq compatible lentivirus was constructed, similar to recent reports. Each barcode consists of a 30nt region with optimal GC content at 100,000-fold diversity. An 8nt sublibrary index was also incorporated to allow flexible control of the overall library size. Upon successful transduction, the lentiviral vector is stably integrated into the genome, resulting in constitutive expression of selection markers, puromycin resistance, and eGFP, as well as a barcode embedded in the 3' untranslated region of the reporter cassette (Figure 17B). To confirm our ability to recover these transduced barcodes, to ensure the system can distinguish the underlying biology, and to evaluate whether dual puromycin and eGFP selection is required, unique barcodes were transduced into five different cell lines (PC9, MCF-10A, MDA-MB-231, NCI-H358, and NCI-H1373). Transduced cells were selected with puromycin alone, mixed together, and profiled by 10× scRNA-seq for a composite mixture of five cell lines. Individual cell expression profiles were clustered and projected into two-dimensional space using UMAP. TraCe-seq barcodes were independently mapped and annotated for each cell. Unsupervised clustering of the five cell lines resulted in five distinct transcriptome clusters, with 81% of TraCe-seq barcodes correctly assigned to the correct transcriptome cluster (Figure 20A). Marker genes from unsupervised scRNA-seq differential gene expression analysis of each TraCe-seq barcode population were consistent with gene expression using bulk RNA-seq for each cell line (Figures 20B, 20C). Overall, barcode detection and mapping varied as a function of their expression (Figures 20D, 20E). Cells with lower average barcode expression had a higher barcode loss rate, which followed the same loss pattern as other endogenously expressed genes (Figure 20D).To overcome this limitation, fluorescence-activated cell sorting (FACS) of the top 50% of eGFP-expressing cells significantly enhanced barcode expression and increased the barcode recovery rate to over 90% (Figure 20D). In summary, a dual-selection marker system using TraCe-seq vectors is necessary to ensure optimal barcode detection and assignment.

[0117] Since EGFR-targeted therapy is a mainstream of precision tumor therapy, we used TraCe-seq to better understand the response to EGFR therapy using a unique MOA. Despite initial treatment success, most patients with common EGFR inhibitor-sensitizing mutations (L858R or exon 19 deletion) eventually develop resistance to EGFR-targeted therapy, and the majority of patients lack further secondary site EGFR mutations when treated with the current standard of care EGFR inhibitor osimertinib. Small molecule strategies targeting EGFR have focused on inhibiting its enzymatic activity, but little is known about the potential consequences of EGFR degradation, as opposed to enzymatic inhibition. For this purpose, we developed GNE-104 (a heterobifunctional degrader consisting of the EGFR inhibitor erlotinib ligated to the VHL binding portion, the substrate-binding component of the CRL4-VHL E3 ligase complex). GNE-104 induced significant dose-dependent degradation of EGFR and potently suppressed phospho-EGFR levels (Figure 17C and Figure 21A). In an in vitro kinase inhibition assay (Figure 21D), GNE-069 was produced as a non-degrader control that maintained a similar potency and selectivity profile to GNE-104, but did not bind to VHL (due to specific stereochemistry at the VHL binding site) and did not degrade EGFR (Figures 21B, 21C) (Figure 17C).

[0118] Next, the inventors investigated whether TraCe-seq could: (1) distinguish between drug-sensitive and resistant clones when treated with specific molecules; (2) capture known genes associated with EGFR inhibitor resistance; and (3) distinguish between the response and resistance mechanisms of EGFR degraders to conventional EGFR kinase inhibitors. PC9 cells were selected as a model system because they respond strongly to EGFR inhibition and contain a well-established subpopulation of resistant cells driven by non-genetic mechanisms. To determine whether it is possible to effectively and reproducibly capture the spectrum of EGFR inhibitor resistance / sensitivity phenotypes in PC9 cells, the inventors first conducted a pilot study on 500 TraCe-seq barcoded clones. These clones were grown in 12 doubling cycles, and 15,000 cells from the resulting population were treated with erlotinib or GNE-104 in replicas for 2 months, and barcode enrichment patterns were determined from the genomic DNA of the surviving population. Overall, clonal survival was remarkably consistent under individual treatment conditions, differing between erlotinib treatment and GNE-104 treatment (Figure 22A). Therefore, it was concluded that this system faithfully captures the spectrum of responses and resistance to EGFR inhibition in PC9 cells under similar conditions.

[0119] To determine the appropriate timing for tracking and comparing immediate responses to EGFR kinase inhibitors versus degraders, the dynamics of treatment responses were characterized over 8 days. Paradoxically and interestingly, despite promoting both kinase inhibition and EGFR degradation, GNE-104 was observed to be far less efficient in inhibiting cell proliferation compared to erlotinib and GNE-069 using multiple assay formats (Figures 22B-22D). Based on the dynamics of the response to the drugs, it was decided to sample the population on day 4, when significant proliferation inhibition had occurred but before susceptible cells had completely disappeared (Figure 22C), thereby capturing this and allowing for a comparison of transcriptional responses between drug-sensitive and drug-resistant populations.

[0120] To perform the TraCe-seq experiment, 600 PC9 cells with unique barcodes were randomly selected, and these cells were grown in approximately 12 doubling cycles to establish a starting population. Single-cell transcriptional profiling was performed on a fraction of this population with approximately 30x barcode coverage to record the baseline transcriptional program and relative clonal abundance. These cells were seeded at 200K and treated with either erlotinib, GNE-104, or GNE-069. Cells were harvested on day 4 and profiled by 10× scRNA-seq to determine the relative compatibility of each clone and capture the transcriptional changes induced by the treatment (Figure 17D). To confirm that the TraCe-seq approach accurately measures relative clonal size at baseline (inferred from TraCe-seq barcode abundance), 20K barcoded cells were grown in DMSO for 7 days and profiled by saturating barcode analysis from genomic DNA (>3000× coverage). In fact, clonal abundance determined by scRNA-seq versus genomic DNA analysis was highly correlated (Figure 23A). 551 barcodes were recovered from baseline before treatment (Figure 17E). All treatment conditions resulted in a decrease in barcode diversity (-16.0 ± 5%). Relative clonal abundance was very similar under erlotinib and GNE-069 treatment, reflecting their very comparable activity (which will hereafter be grouped together under the term kinase inhibitors), but GNE-104 treatment resulted in a more diverse pattern (Figure 23B). Consistent with the observed decrease in proliferation inhibitory activity of GNE-104 (Figures 22B-22D), GNE-104 treatment was not effective in reducing both the absolute number and diversity of TraCe-seq barcodes (Figure 17E). Furthermore, while comparable MAPK pathway suppression was observed across all treatment conditions (Figure 17F), erlotinib and GNE-069 induced significantly greater cell cycle arrest (G0) than GNE-104 (p<10). -145(Figure 17G) GNE-104 treatment suggested that more cells could persist in the presence of MAPK pathway inhibition. Reduced antiproliferative effect of GNE-104 was observed in several EGFR-mutated lung cancer cell lines (Figures 24A and 24B). Since 10-fold higher concentrations of free VHL ligand did not affect the response to erlotinib or GNE-069, this loss of efficacy was not attributable to the presence of active VHL ligand (Figure 24C).

[0121] To further understand the differential responses of PC9 cells to these treatments at the clonal level, we attempted to classify Trace-seq barcodes that were overexpressed or underexpressed in the treatment population compared to the start population (Figure 23C). Overexpressed Trace-seq barcodes under treatment compared to the untreated population were classified into three classes of resistant clones (Figure 18A), and underexpressed barcodes under treatment were classified into three classes of susceptible clones based on their response to kinase inhibitors versus degraders (Figure 23D). In particular, GNE-104 treatment resulted in the overexpression of a unique set of Trace-seq barcodes compared to kinase inhibitors categorized as degrader-resistant (Figure 18A). It was hypothesized that overexpressed and underexpressed clones possess differential baseline features that cause cells to respond or persist under subsequent treatments. To investigate this, we attempted to identify differences in gene expression between drug-susceptible and drug-resistant clones at baseline under untreated conditions. To search for existing transcriptional features that may confer resistance to kinase inhibitors (erlotinib and GNE-069), we performed unbiased differential gene expression analysis by comparing baseline gene expression between kinase inhibitor-resistant clones and kinase inhibitor-sensitive clones. Consistent with previous reports, VIM and AXL were found to be significantly information-regulated in kinase inhibitor-resistant cells before treatment (Figure 18B), confirming that TraCe-seq successfully captures the expected resistance subpopulation and associated molecular markers. Importantly, degrader-resistant clones showed only elevated AXL expression but no upregulation of VIM, suggesting that they are indeed transcriptionally different from kinase inhibitor-resistant clones (Figure 18B). To clarify what transcriptional programs may enable cell survival under degrader treatment, we performed unbiased gene set expression analysis comparing degrader-resistant and degrader-sensitive clones (Table 1).Interestingly, "protein processing in the ER" activity was found to be the most significant and enriched pathway, differing between degrader-sensitive and degrader-resistant clones (Figure 18C), suggesting that cells with lower expression of ER protein processing genes may be protected from GNE-104.

[0122] [Table 1]

[0123] To further investigate adaptive transcriptional programs associated with treatment resistance to kinase inhibitors or degraders, trajectory inference analysis was performed by systematically ordering treated cells over sham time and tracking the evolution of cell state in an unsupervised manner. This analysis identified four pathways (a-d) (Figures 18D-18F) with different processes and destinies. The distribution of kinase inhibitor-sensitive versus kinase inhibitor-resistant clones suggested that pathways (b) and (c) were associated with kinase inhibitor resistance, while pathway (d) represented sensitivity (Figures 25A-25B). Consistently, VIM expression was elevated along pathways (b) and (c), representing general resistance to EGFR inhibition (Figure 18F), while along pathway (d), MAPK pathway activity scores were most strongly suppressed, G0 / quiescent cell state was most elevated, and this was consistent with sensitivity. In contrast to the distribution of kinase inhibitor-resistant clones, degrader-resistant clones were found to be most dominant along pathway (a) instead (Figure 18G). Furthermore, these clones were prominently distributed at the end of pathway (a) under GNE-104 treatment, but their presence increased in the sensitivity trajectory (d) under kinase inhibitor treatment (Figure 25C). Interestingly, genes related to protein processing in the ER pathway were uniquely downregulated along pathway (a) (Figure 18F), suggesting that reduced expression of ER protein processing genes is uniquely associated with the treatment adaptation and survival of these cells under EGFR degrader treatment.

[0124] To investigate whether reduced expression of ER protein processing genes is a key feature of the adaptive transcriptional response in degrader-resistant clones treated with GNE-104 compared to kinase inhibitors, we performed unbiased gene set expression analysis. Indeed, protein processing in the ER pathway was revealed to be the most enriched and significant pathway induced by erlotinib treatment compared to GNE-104 treatment in these degrader-resistant clones (Figure 18H, Table 2). In summary, this suggests that cells with lower pre-existing expression of ER protein processing genes are more prepared to resist EGFR degraders; the inability of degraders to activate these ER protein processing genes to the levels observed with kinase inhibitors may underlie the paradoxical loss of efficacy with our dual inhibitor-degrader EGFR targeting agent.

[0125] [Table 2]

[0126] A clear difference between kinase inhibitor and degrader treatment is the sustained presence of EGFR protein. To test whether the presence of EGFR protein explains these differences, siRNA was used in combination with kinase inhibitor treatment to halt EGFR protein production (Figure 26A). EGFR knockdown alone resulted in substantial loss of EGFR protein (Figure 26B) and strong MAPK pathway suppression (Figures 26C, 26D), while siRNA treatment was far less effective in killing PC9 and HCC4006 cells (Figures 19A, 26E). Furthermore, cells treated with siEGFR were paradoxically protected from treatment with the kinase inhibitors erlotinib and osimertinib (Figure 19A, 26D). These results further suggest that a decrease in EGFR protein levels may paradoxically impair the cellular efficacy of EGFR kinase inhibitors.

[0127] To further test this hypothesis, the EGFR degrader GNE-641 was constructed based on the allosteric EGFR ligand EAI-045 linked to a VHL-binding ligand (Figures 27A, 27B). Based on molecular modeling (data not shown) and recent reports, EAI-045 was predicted to be able to bind to EGFR L858R concurrently with osimertinib. Consistent with modeling predictions, EGFR was efficiently degraded in cells co-treated with osimertinib and GNE-641 (Figure 19B). As before, an inactive degrader control was also generated that did not affect EGFR protein levels (Figure 19B) (GNE-640, Figures 27A, 27B). Despite its ability to degrade EGFR, GNE-641 exhibited very modest anti-growth activity on its own (Figure 27C). Furthermore, it is noteworthy that co-treatment with GNE-641 and osimertinib promoted the survival of both NCI-1975 and NCI-H3255 cells, despite the potent biochemical activity and apparent effects on MAPK pathway target genes (Figure 19B, Figure 26G) (compared to osimertinib alone or co-treatment with osimertinib and inactive GNE-640, Figure 27D, Figure 27E). In summary, these results strongly suggest that the EGFR protein itself plays a crucial role in mediating the cellular efficacy of EGFR kinase inhibitors.

[0128] EGFR is a transmembrane protein that transports the vesicle system (including the ER) throughout its life cycle. Considering that both high expression of protein processing genes in the ER and persistent presence of EGFR protein are strongly associated with the full efficacy of EGFR kinase inhibitors, we hypothesized that a death-promoting signal originating from the ER compartment may be responsible for the enhanced efficacy of EGFR inhibitors (compared to degraders). Specifically, disruption of the ER proteasis, such as increased protein misfolding and secretory overload, is known to increase ER stress. When ER stress exceeds a certain threshold, it triggers effective death-promoting signals via the ISR PERK-ATF4-CHOP axis (Figure 28A). We hypothesized that the persistent presence of EGFR protein bound to a kinase inhibitor may lead to activation of the downstream PERK-ATF4-CHOP death-promoting signaling cascade and thus contribute to cytotoxicity. In contrast, EGFR degraders actively remove inhibitor-bound EGFR proteins, thus minimizing ER stress induced by inhibitor-bound EGFR (Figure 19C). Indeed, erlotinib and osimertinib were found to significantly induce the expression of key downstream cell death-promoting ISR genes (ATF4, CHOP, GADD34) in PC9, HCC4006, and NCI-H1975 cells, and to significantly induce the major transcriptional target of ATF4 (SLC7A5) in PC9 and NCI-H1975 cells (Figures 19D, 19E, 26F). Furthermore, such induction was blunted upon acute depletion of EGFR protein production by siEGFR (Figures 19D, 26F) or allosteric degraders (Figure 19E).

[0129] To further investigate the role of ER stress in the efficacy of EGFR kinase inhibitors, two questions are asked: 1) Does blocking ATF4 induction conflate resistance to EGFR kinase inhibitors? 2) Does pharmacological induction of ER stress compensate for the lost efficacy of EGFR degraders? To specifically block the excessive activation of death-promoting genes downstream of ER stress, cells were treated with an ISR inhibitor (ISRIB) for 24 consecutive hours, followed by the addition of an EGFR kinase inhibitor. ISRIB treatment did indeed blunt the activation of ATF4, CHOP, and GADD34 induced by EGFR kinase inhibitors (Figures 28A, 28B, and 28D), and reduced cell death in both PC9 and H1975 cells (Figures 19F, 28C, and 28F). Conversely, when ER stress inducers such as tunicamycin or thapsigargin (Figure 28A) were added at very low, non-toxic concentrations, the death-promoting ER stress gene was greatly induced in PC9 cells, and the cytotoxic activity of the degrader GNE-104 was enhanced (Figures 28E-28G).

[0130] Inspired by these findings, we investigated whether direct activation of ISRs could further enhance the efficacy of clinically approved EGFR inhibitors. We used the PERK activator CCT 020312, which specifically activates the PERK-ISR branch downstream of ER stress, compared to the broader ER stress inducers tunicamycin and thapsigarzine. When administered in combination with erlotinib or osimertinib, CCT 020312 further enhanced erlotinib and osimertinib-induced ISR gene induction (Figures 29A-29C), enhanced cell death induction (Figure 29D), and effectively eliminated residual cells commonly present after acute EGFR inhibitor treatment (Figures 19G-19I). These results highlight previously unrecognized means of enhancing responses to anti-EGFR therapies of clinical interest.

[0131] In summary, the TraCe-seq platform enables the identification of existing adaptive transcriptional features that influence the outcomes of different therapeutic treatments. By tracking responsive and sustained cells after exposure to different EGFR-targeted therapies and performing differential genetic analysis in combination with spurious time trajectory comparisons, unexpected differences in cellular mechanisms between EGFR targeters with differential MOAs were revealed. This revealed a stimulating and previously unrecognized component underlying the EGFR inhibitor response: induction of ER stress by inhibitor-bound EGFR is essential for achieving full efficacy. A detailed understanding of the biochemical mechanisms underlying inhibitor-bound EGFR and ER stress induction could aid in the future development of small molecules against EGFR and possibly other membrane-related proteins. Furthermore, the results challenge the broad assumption that targeted degradation is universally superior to occupation-based inhibition and suggest that the underlying biology should be carefully considered when selecting compound MOAs.

[0132] The TraCe-seq approach will guide the development of future therapies by revealing unknown features that predict responses and resistances to different molecular modalities or combinations of treatments. Furthermore, because barcodes are stably integrated and constitutively excluded, TraCe-seq barcoded populations can be sampled at multiple time points to collect transcriptional information at clonal resolution over time, thus providing a generally applicable approach for studying the evolution of heterologous populations under various circumstances. Moreover, further modifications of the method that allow for the direct isolation of living cells within existing subpopulations of interest can further expand its use across diverse systems.

[0133] method Cell lines and tissue cultures: Cell lines were obtained, characterized, and quality-controlled as described. Cell lines were maintained using standard tissue culture techniques. All cell lines (except MCF-10A) were cultured in RPMI containing 10% thermo-inactivated fetal bovine serum and 2 mM L-glutamine. MCF10-A cells were cultured in DMEM-F12 (Thermo Fisher Scientific, 11330-032), 5% horse serum (Sigma Aldrich, H1270), 10 μg / ml insulin (Sigma Aldrich, I-1882), 500 ng / ml hydrocortisone (Sigma Aldrich, H-0888), 100 ng / ml cholera toxin (Sigma Aldrich, C-8052), and 20 ng / ml EGF (Peprotech, 100-15R).

[0134] Lentivirus Production and Cell Transduction: A barcode library of TraCe-Seq lentiviruses was synthesized by Cellecta Inc. Lentiviruses were produced in 293T cells by co-transfecting the barcode library with pCMVdR8.9 (expressing gag, pol, and rev genes) and pCMV-VSV-g (expressing envelope protein). The viruses were enriched using Lenti-X-Concentrator and stored at -80°C. For viral transduction, 5 × 10⁶ cells were used. 6Cells were seeded in T75 flasks and infected overnight with 8 μg / ml polyblen (TR-1003-G, EMD Millipore) at an MOI (Multiplicity of Infection) of 0.05–0.1. Cells were initially divided two days after viral infection. To determine the MOI, a fraction of cells was plated in 96-well plates at 5,000 cells / well in 100 μl of medium. Cells were then incubated with or without 2 μg / ml puromycin, and the percentage of infected puromycin-resistant cells was determined by a viability assay using CellTiter-Glo Luminescent Cell Viability Assay (Promega catalog number G7572) to confirm that the infection MOI was within the range. The remainder of the cells were grown in 2 μg / ml puromycin.

[0135] A TraCe-seq library enriched with eGFP-expressing cells by FACS was introduced, selected cells were dissociated into single cells with puromycin, and then placed in PBS in a 10x10 format. 6 The cells were resuspended at / ml. The top 50% of eGFP-expressing cells were collected using a BD Aria Fusion cell separator and grown in cell culture medium containing 2 μg / ml puromycin.

[0136] Single-cell RNA sequencing: Cultured cells were trypsin-treated into single-cell suspension and processed using a chromium single-cell gene expression 3' library and gel bead kit according to the manufacturer's instructions (10X Genomics, Pleasanton, CA). Cells were counted and viability checked using a Vi-CELL XR cell counter (Beckman Coulter, Brea, CA), then injected into a microfluidic chip to form gel beads in emulsion (GEM) using a 10X chromium instrument. Reverse transcription was performed on the GEMs, and the RT products were purified and amplified. Expression libraries were prepared from cDNA, profiled using the Bioanalyzer High Sensitivity DNA Kit (Agilent Technologies, Santa Clara, CA), and quantified using the Kapa Library Quantification Kit (Kapa Biosystems, Wilmington, MA). Libraries were sequenced using Illumina HiSeq2500 or HiSeq4000 (Illumina, San Diego, CA).

[0137] chemical synthesis VHL Ligands: The VHL ligand (GNE-429) used in the competitive assay in Figure 21B was synthesized as described above. The active and inactive VHL ligands (GNE-128 and GNE-127, respectively) used in Figures 17C and 27A were synthesized as described above. Note that the stereochemistry of these two compounds was assigned based on biochemical potency (referenced in the literature), with GNE-128 as the active isomer.

[0138] Erlotinib-derived degrader:

[0139] 4-((3-ethynylphenyl)amino)-7-(2-methoxyethoxy)quinazolin-6-ol

[0140] [ka]

[0141] 4-((3-ethynylphenyl)amino)-7-(2-methoxyethoxy)quinazolin-6-ol was prepared in the same manner as described in (4) above. As a final step, a solution of 7-(2-methoxyethoxy)-4-((3-((trimethylsilyl)ethynyl)phenyl)amino)quinazolin-6-ol (1.23 g, 3.02 mmol) and TBAF (6 mL, 1 M in THF) in THF (10 mL) was stirred at room temperature for 20 minutes. The solution was concentrated under vacuum. The residue was purified by silica gel flash chromatography eluted with ethyl acetate / petroleum ether (0-100%) to obtain 560 mg (yield 55%) of the title compound as a pale yellow solid. LCMS(ESI):[M+H] + =336. 1 H NMR(300MHz,DMSO-d6)δ9.65(s,1H),9.43(s,1H),8.47(s,1H),8.08(t,J=1.9Hz,1H),7.92-7.85(m,1H),7.80(s,1H), 7.37(t,J=7.9Hz,1H),7.20(s,1H),7.18-7.12(m,1H),4.36-4.26(m,2H),4.19(s,1H),3.83-3.73(m,2H),3.35(s,3H).

[0142] (2S,4S)-1-((S)-2-amino-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazole-5-yl)benzyl)pyrrolidine-2-carboxamide hydrogen chloride

[0143] [ka]

[0144] (2S,4S)-1-((S)-2-amino-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazole-5-yl)benzyl)pyrrolidine-2-carboxamide hydrogen chloride was prepared using 1-tert-butyl2-methyl(2S,4S)-4-hydroxypyrrolidine-1,2-dicarboxylate as the starting material (2), similar to the preparation described for (2S,4R)-1-[(2S)-2-amino-3,3-dimethylbutanoyl]-4-hydroxy-N-[[4-(4-methylthiazole-5-yl)phenyl]methyl]pyrrolidine-2-carboxamide. In the final step, a solution of tert-butyl N-[(2S)-1-[(2S,4S)-4-hydroxy-2-([[4-(4-methyl-1,3-thiazole-5-yl)phenyl]methyl]carbamoyl)pyrrolidine-1-yl]-3,3-dimethyl-1-oxobutan-2-yl]carbamate (718 mg, 1.35 mmol) in dichloromethane (6 mL) and HCl / 1,4-dioxane (3 mL, 4 M) is prepared in 25 o The mixture was stirred in 1C for 0.5 hours. The resulting mixture was concentrated under vacuum. The residue was stirred with ethyl ether (5 mL). The solid was recovered by filtration and dried under vacuum to obtain 700 mg (crude) of the title compound as a white solid. LCMS(ESI):[M+H] + =431.3.1H NMR(300MHz,DMSO-d6)δ9.04(s,1H),8.79(t,J=6.0Hz,1H),8.16(s,3H),7.41(s,4H),4.52-4.18(m,5H),4.02(dd,J=10.3,5.9Hz,1 H),3.94(d,J=5.5Hz,1H),3.31(dd,J=10.2,6.6Hz,1H),2.45(s,3H),2.42-2.30(m,1H),1.74(dt,J=12.5,7.3Hz,1H),1.03(s,9H).

[0145] Synthetic scheme for the preparation of ethyl 2-((6-(2-bromoethoxy)hexa-2,4-diyn-1-yl)oxy)acetate

[0146] [ka]

[0147] Diethyl 2,2'-(hexa-2,4-diyne-1,6-diylbis(oxy))diacetate

Chem.

[0148] Sodium hydride (1.08 g, 45.0 mmol) was added to a solution of hexa-2,4-diyne-1,6-diol (1 g, 9.08 mmol) in tetrahydrofuran (20 mL) under nitrogen. The mixture was stirred at 25 o °C for 1 hour under a nitrogen atmosphere. Then, ethyl 2-bromoacetate (4.50 g, 26.9 mmol) was added, and the resulting solution was stirred at 25 o °C for 3 hours. It was concentrated under vacuum. The residue was purified by flash chromatography on a silica gel column eluting with ethyl acetate / petroleum ether (0%-50%) to give 1.3 g (51%) of the title compound as a colorless oil. LCMS (ESI): [M+18] + = 300.

[0149] Ethyl 2-((6-(2-hydroxyethoxy)hexa-2,4-diyne-1-yl)oxy)acetate

Chem.

[0150] LiBH4 (86.0 mg, 3.95 mmol) was added to a solution of ethyl 2-[[6-(2-ethoxy-2-oxoethoxy)hexa-2,4-diyne-1-yl]oxy]acetate (1.10 g, 3.90 mmol) in tetrahydrofuran (10 mL) at 0 o °C. The resulting solution was stirred at 25 oThe mixture was stirred with C. The reaction product was quenched by the addition of ethyl acetate. The resulting mixture was concentrated under vacuum. The residue was partitioned between ethyl acetate and water. The organic layer was dried over anhydrous sodium sulfate and concentrated under vacuum. The residue was purified by flash chromatography on a silica gel column eluted with ethyl acetate / petroleum ether (0%-80%) to obtain 237 mg (25%) of the title compound as a colorless oil. LCMS(ESI):[M+H] + =241.

[0151] Ethyl 2-((6-(2-((methylsulfonyl)oxy)ethoxy)hexa-2,4-diyne-1-yl)oxy)acetate [ka]

[0152] A solution of ethyl 2-[[6-(2-hydroxyethoxy)hexa-2,4-diyn-1-yl]oxy]acetate (237 mg, 0.986 mmol), methanesulfonylmethanesulfonate (348 mg, 2.00 mmol), and DIPEA (516 mg, 3.99 mmol) in dichloromethane (4 mL) was prepared in 25 ml. o The mixture was stirred in 1°C for 1 hour. The resulting mixture was washed with water, dried, and concentrated. This yielded 320 mg (crude) of the title compound as a brown oily substance. LC-MS(ESI):[M+H] + =319. The crude material was used in the next process without further refinement.

[0153] Ethyl 2-((6-(2-bromoethoxy)hexa-2,4-diin-1-yl)oxy)acetate [ka]

[0154] A solution of ethyl 2-([6-[2-(methanesulfonyloxy)ethoxy]hexa-2,4-diyne-1-yl]oxy) acetate (320 mg, 1.005 mmol) and KBr (354 mg, 2.98 mmol) in N,N-dimethylformamide (3 mL) was prepared in 60 ml. o The mixture was stirred overnight in 1C. It was then cooled to room temperature. The reaction mixture was partitioned between ethyl acetate and water. The organic layer was dried over anhydrous sodium sulfate and concentrated under vacuum. The residue was purified by flash chromatography on a silica gel column using ethyl acetate / petroleum ether (0%-40%) to obtain 222 mg (73%) of the title compound as a colorless oil. LC-MS (ESI): [M+H] + =303 / 305.1H NMR(300MHz,DMSO-d6)δ4.43-4.34(m,4H),4.19-4.06(m,4H),3.82-3.73(m,2H),3.66-3.56(m,2H),1.20(t,J=7.1Hz,3H).

[0155] Synthetic scheme for the preparation of GNE-104, (2R,4R)-1-((S)-2-(2-((6-(2-((4-((3-ethynylphenyl)amino)-7-(2-methoxyethoxy)quinazoline-6-yl)oxy)ethoxy)hexa-2,4-diin-1-yl)oxy)acetamide)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazole-5-yl)benzyl)pyrrolidine-2-carboxamide. [ka]

[0156] Ethyl 2-((6-(2-((4-((3-ethynylphenyl)amino)-7-(2-methoxyethoxy)quinazoline-6-yl)oxy)ethoxy)hexa-2,4-diin-1-yl)oxy)acetate [ka]

[0157] A solution of 4-(3-ethynylanilino)-7-(2-methoxyethoxy)quinazolin-6-ol (50.0 mg, 0.150 mmol), ethyl 2-[6-(2-bromoethoxy)hexa-2,4-diinoxy]acetate (67.8 mg, 0.220 mmol), and K2CO3 (41.2 mg, 0.300 mmol) in N,N-dimethylformamide (2 mL) was prepared in 60 ml. o The mixture was stirred in 1C for 2 hours. The reaction solution was loaded onto a reverse-phase column for direct elution with CH3CN / H2O (0.5% NH4HCO3) (0-100%) to obtain the title compound (46 mg, yield 55.3%) as a white solid. LC-MS (ESI): [M+H] + =558

[0158] 2-((6-(2-((4-((3-ethynylphenyl)amino)-7-(2-methoxyethoxy)quinazoline-6-yl)oxy)ethoxy)hexa-2,4-diyne-1-yl)oxy)acetic acid [ka]

[0159] A solution of ethyl 2-[6-[2-[4-(3-ethynylanilino)-7-(2-methoxyethoxy)quinazolin-6-yl]oxyethoxy]hexa-2,4-diinoxy]acetate (46.0 mg, 0.0800 mmol) and LiOH (5.94 mg, 0.250 mmol) in tetrahydrofuran (2 mL) and water (1 mL) was prepared in 25 o The mixture was stirred in C for 1 hour. The filtrate was concentrated under vacuum. The residue was purified by reverse-phase column elution with CH3CN / H2O (0.5% NH4HCO3) (0-100%) to obtain the title compound (40 mg, 91.6% yield) as a white solid. LCMS(ESI):[M+H] + =530

[0160] GNE-104, (2R,4R)-1-((S)-2-(2-((6-(2-((4-((3-ethynylphenyl)amino)-7-(2-methoxyethoxy)quinazoline-6-yl)oxy)ethoxy)hexa-2,4-diin-1-yl)oxy)acetamide)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazole-5-yl)benzyl)pyrrolidine-2-carboxamide [ka]

[0161] 2-[6-[2-[4-(3-ethynylanilino)-7-(2-methoxyethoxy)quinazolin-6-yl]oxyethoxy]hexa-2,4-diinoxy]acetic acid (40.0 mg, 0.080 mmol), (2S,4R)-1-[(2S)-2-amino-3,3-dimethylbutanoyl]-4-hydroxy-N-[[4-(4-methylthiazole-5-yl)phenyl]methyl]pyrrolidine-2-carboxamide (41.1 mg, 0.080 mmol), and DIPEA (10.3 mg, 0.080 mmol) were dissolved in N,N-dimethylformamide (2 mL) to which HATU (28.7 mg, 0.080 mmol) was added. The resulting solution was 25 o The solution was stirred in 1°C for 1 hour. The solution was loaded onto a reverse-phase column for direct elution with CH3CN / H2O (0.1% FA) (5-100%) to obtain the title compound (29 mg, yield 40.8%) as a white solid. LC-MS (ESI): [M+H] + =942. 1H NMR(400MHz,DMSO-d6)δ9.47(s,1H),8.98(s,1H),8.60(t,J=6.1Hz,1H),8.50(s,1H),7.99(t,J =1.9Hz,1H),7.93-7.84(m,2H),7.56(d,J=9.4Hz,1H),7.48-7.36(m,5H),7.26-7.17(m,2H),5. 15(d,J=3.6Hz,1H),4.56-4.18(m,14H),4.03(s,2H),3.95-3.88(m,2H),3.78-3.71(m,2H),3.7 0-3.56(m,2H),3.36(s,3H),2.44(s,3H),2.06(d,J=8.9Hz,1H),1.96-1.84(m,1H),0.94(s,9H).

[0162] GNE-069, (2R,4S)-1-((S)-2-(2-((6-(2-((4-((3-ethynylphenyl)amino)-7-(2-methoxyethoxy)quinazoline-6-yl)oxy)ethoxy)hexa-2,4-diin-1-yl)oxy)acetamide)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methylthiazole-5-yl)benzyl)pyrrolidine-2-carboxamide [ka]

[0163] HATU (47.4 mg, 0.120 mmol) was added to a solution of 2-[6-[2-[4-(3-ethynylanilino)-7-(2-methoxyethoxy)quinazolin-6-yl]oxyethoxy]hexa-2,4-diinoxy]acetic acid (60.0 mg, 0.110 mmol), (4S)-1-[(2S)-2-amino-3,3-dimethylbutanoyl]-4-hydroxy-N-[[4-(4-methylthiazole-5-yl)phenyl]methyl]pyrrolidine-2-carboxamide hydrochloride (52.9 mg, 0.110 mmol), and DIPEA (43.9 mg, 0.340 mmol) in N,N-dimethylformamide (2 mL). The resulting solution was stirred at room temperature for 1 hour. The reaction solution was loaded onto a reverse-phase column eluted with CH3CN / H2O (0.5% NH4HCO3) (0-100%) to obtain the title compound (59.6 mg, yield 55.8%) as a pale yellow solid. LC-MS (ESI): [M+H] + =942. 1 H NMR(400MHz,DMSO-d6)δ9.47(s,1H),8.98(s,1H),8.66(t,J=6.0Hz,1H),8.50(s,1H),7.99(t,J=1.9Hz, 1H),7.93-7.85(m,2H),7.58(d,J=9.0Hz,1H),7.40(d,J=3.8Hz,5H),7.26-7.17(m,2H),5.44(d,J=7.2Hz ,1H),4.54-4.17(m,14H),4.08-3.95(m,2H),3.90(dt,J=16.9,5.2Hz,3H),3.75(t,J=4.4Hz,2H),3.45(d d,J=10.1,5.4Hz,1H),3.35(s,3H),2.44(s,3H),2.35(m,1H),1.74(dt,J=12.5,6.1Hz,1H),0.95(s,9H).

[0164] The synthesis procedures for the allosteric-derived degraders GNE-640 and GNE-641 are detailed in International Publication No. 2019183523 (identified as 1002.3 and 1002.4, respectively). Briefly, the desired stereochemistry of the EGFR allosteric ligand was established from the crystal structure of the associated ligand EAI-001 bound to EGFR (T790M / V948R). The (R) stereochemistry assigned to GNE-640 and GNE-641 (Figure 27A) was inferred from relative affinity.

[0165] Compound dosage matching: GNE-069 and GNE-104 were applied at an equimolar concentration of 1 μM because the two compounds are identical in terms of their anti-EGFR specificity, potency, and physicochemical properties. The dose of erlotinib was empirically determined by matching the elimination efficiency and pharmacokinetics with those induced by 1 μM GNE-069 (Figure 22C).

[0166] Clonality assay: Cells were seeded at the desired density in 6-well plates using 2-3 replications and allowed to adhere overnight. The following day, cells were treated with the indicated compounds. Medium containing the appropriate compounds was replenished every 48-72 hours. At the end of the assay, cells were washed once with PBS, fixed, stained with crystal violet solution (Sigma Aldrich, HT90132) for 20 minutes, washed with water, and dried before scanning. For the experiments shown in Figure 18E, the erlotinib concentration was determined to match the cell viability effect of 1 μM GNE-069 in each cell line. As a result, 1 μM erlotinib was applied to PC-9 and HCC2935, and 100 nM erlotinib was applied to HCC4006 and HCC827. For the experiment shown in Figure 27B, cells were treated with either DMSO or 1 μM ISRIB (Tocris catalog number 5284) for 24 hours, followed by co-treatment with an EGFR inhibitor for 8 days. For the other experiment, cells were treated with 1 μM GNE-104 for 24 hours, followed by co-treatment with 0.5 μg / ml tunicamycin (Sigma Aldrich, T7765) or 1 nM thapsigargin (Sigma Aldrich, T9033) for 3 days.

[0167] Quantification of cell viability and cell death: For the experiments shown in Figures 24C and 27D, the indicated cells were seeded at the desired density in 24-well plates using replication and allowed to adhere overnight. The following day, the cells were treated with the indicated compounds. Medium containing the appropriate compounds was replenished every 48 hours. At each medium change, cell viability was quantified using the CellTiter-Glo Luminescent Cell Viability Assay (Promega, G7572). For the experiments shown in Figures 19I and 29D, PC9 cells were seeded at the desired density in 96-well plates and allowed to adhere overnight. The following day, the cells were treated with the indicated compounds or combinations of compounds at the indicated concentrations. Induction of cell death was quantified 24 hours after treatment using the Caspase-Glo 3 / 7 Assay System (Promega, G8093), and cell viability was measured 72 hours after treatment using the CellTiter-Glo Luminescent Cell Viability Assay (Promega, G7572). CCT020312 was obtained from EMD Millipore (324879).

[0168] Immunoblotting: Immunoblotting was performed using a standard method. Cells were briefly washed with ice-cold PBS and lysed with the following lysis buffer (1% NP40, 50 mM Tris, pH 7.8, 150 mM NaCl, 5 mM EDTA) + protease inhibitor mixture (complete mini tablets, Roche Applied Science, 11836170001) and phosphatase inhibitor mixture (Thermo Fisher Scientific, 78420). The lysates were 4 oThe proteins were centrifuged at 15,000 rpm for 10 minutes in 1C, and the protein concentration was determined by BCA (Thermo Fisher Scientific, 23227). Equal amounts of protein were separated by SDS-PAGE on NuPAGE, 4-12% Bis-Tris gel (Thermo Fisher Scientific, WG-1403) and transferred to a nitrocellulose membrane (Bio-Rad, 170-4159). The membrane was blocked in blocking buffer (Li-Cor, 927-40000), incubated overnight with the indicated primary antibody, and analyzed by adding the secondary antibody IRDye 680LT goat anti-mouse IgG (Li-Cor, 926-68050) or IRDye 800CW goat anti-rabbit IgG (Li-Cor, 926-32211). The membrane was visualized using a Li-Cor Odyssey CLx scanner. Anti-EGFR (MI-12-1) was purchased from Medical and Biological Labs, while anti-pEGFR (3777), anti-β-tubulin (2146), and anti-β-actin (4970) were purchased from Cell Signaling Technology. The IR-conjugated secondary antibodies goat anti-mouse 680LT (926-68020) and goat anti-rabbit 800CW (926-32211) were purchased from Li-Cor.

[0169] RNA extraction and qRT-PCR: RNA was extracted using the Qiagen RNeasy Plus kit according to the manufacturer's instructions. The RNA was reverse transcribed using the High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific catalog no. 4368814). qPCR was performed using TaqMan® Fast Advanced Master Mix (Thermo Fisher Scientific catalog no. 4444557) on an ABI QuantStudio 7 Flex Real-Time PCR System. Relative gene expression levels were quantified using HPRT1 as an internal control. Taqman probes were obtained from Thermo Fisher Scientific (catalog no. 431182, 4448490). These probes were used in the study: ATF4 Hs00909569_g1, DDIT3 / CHOP Hs00358796_g1, PPP1R15A / GADD34 Hs00169585_m1, SLC7A5 Hs01001183_m1, ETV4 Hs00385910_m1, SPRY4 Hs00540086_m1, DUSP6 Hs04329643_s1, HPRT1 Hs02800695_m1.

[0170] siRNA transfection: PC9 cells were seeded in 6-well plates at 200K / well and allowed to adhere overnight. The following day, cells were transfected with siNTC pool (Dharmacon, Inc. catalog number D-001810-10-50) or siEGFR (Dharmacon, Inc. catalog numbers J-003114-12-0050 and J-003114-13-0050) at a final concentration of 50 nM using Lipofectamin RNAiMAX transfection reagent (Thermo Fisher Scientific, catalog no. 13778030). Two hours after transfection, DMSO, erlotinib (1 μM), or osimertinib (100 nM) was added. The culture medium was then replenished every 48–72 hours.

[0171] scRNA-Seq data processing and analysis: scRNA-Seq data were processed using CellRanger 2.1.0 with mkfastq and counting commands. Expression data were processed using a pre-constructed human reference GRCh38. CellRanger performed default filtering for quality control, and data from all samples were combined using the Seurat package v.3.0.0 to form aggregated Seurat objects using the Seurat best practice workflow. Cells were combined into a single Seurat object, followed by ScaleData using UMI counting and G2M cell cycle scoring (Seurat best practice). FindNeighbors, followed by FindCluster and RunUMAP, were performed on the top 10 PCA components using 2,304 genes identified by FindVariableFeatures. Cell clustering was performed using Luverin at a resolution set to 0.5. Differential expression was performed using FindMarkers and the Wilcoxon rank-sum test. The module score for a given gene set or pathway was calculated cell by cell using the Seurat AddModuleScore function.

[0172] Long-term resistance assay: An experimental population of approximately 500 cells with unique TraCe-seq barcodes was used as the starting population. This population was grown in 12 doubling cycles, and then 15,000 cells (representing 30x barcode coverage) were seeded in each replicate for the long-term resistance study. Erlotinib was administered at the highly stringent dose reported in the literature, and GNE-104 was administered at half the concentration of erlotinib to represent a lower stringent treatment condition. The culture medium was replenished every 48 hours. After 2 months of culture, the cell pellet was sent to Cellecta Inc. for genomic DNA extraction, library preparation, NGS-703 sequencing, and quantification of barcode abundance.

[0173] TraCe-Seq barcode assignment: TraCe-Seq barcodes were recovered from scRNA-Seq FASTQ using Salmon (v.0.1.1). Briefly, a custom enhanced Salmon index was created using human reference GRCh38 and a transgene GFP fused to one of 100,000 30nt GC-optimized barcodes in the library. Next, FASTQ transcriptional expression per isolated cell was quantified using the custom Salmon index. Cells with single GFP barcode expression were assigned to the corresponding TraCe-Seq barcode. Cells with multiple expressed TraCe-Seq barcodes were assigned to a single TraCe-Seq barcode if the expression of one TraCe-Seq was three times higher than that of other detected TraCe-Seq barcodes.

[0174] TraCe-Seq barcode enrichment / depletion analysis: Nonparametric local regression was applied to the occurrence rate of TraCe-Seq barcodes in baseline samples compared to each treatment group. TraCe-Seq barcodes enriched after treatment were those with a local regression of more than 6. Depleted TraCe-Seq barcodes were defined as those with a log2 factor change greater than the mean + standard deviation of the TraCe-Seq population.

[0175] Bulk RNA-seq data processing: For RNA-seq data analysis, RNA-seq reads were first aligned to ribosomal RNA sequences to remove ribosomal reads. The remaining reads were aligned to the human reference genome (GRCh38) using GSNAP version "2013-10-10", allowing for a maximum of two mismatches per 75 base pairs (parameters: '-M 2-n 10-B 2-i 1-N 1-w 200000-E 1--pairmax-rna=200000--clip-overlap'). Transcriptional annotation was based on the Ensembl gene database (release 77). To quantify gene expression levels, the number of reads mapped to exons of each RefSeq gene was calculated.

[0176] Gene set enrichment analysis: KEGG gene signature enrichment was performed using clusterProfiler (v.3.14.3). Gene set enrichment of KEGG signatures was calculated using nPerm=1000 and minGSSIze=120.

[0177] Trajectory Inference Analysis: Erlotinib-treated, GNE-104-treated, and GNE-069-treated cells were subsetted and reclustered using Seurat (see "scRNA-Seq Data Processing and Analysis"). Slingshot 1.5.2 was used to order the cells and project them onto principal curves representing different trajectories. Default Slingshot parameters were used to identify the starting cluster and the number of trajectories / pathways. The length of each trajectory represents spurious time, a calculated parameter indicating progression along the process. For visualization, spurious time was normalized over the entire length of the trajectory to allow comparison of expression trends. References 1. M. Yu et al., A resource for cell line authentication, annotation and quality control. Nature 520, 307-311 (2015). 2.C.Galdeano et al.,Structure-guided design and optimization of small molecules targeting the protein-protein interaction between the von Hippel-Lindau(VHL)E3 ubiquitin ligase and the hypoxia inducible factor(HIF)alpha subunit with in vitro nanomolar affinities.J Med Chem 57,8657-8663(2014). 3.N.Blaquiere et al.,Preparation of(4-hydroxypyrrolodin-2-yl)-hydroxamate compds.And methods of use thereof as modulators of targeted ubiquitination.WO2019084026,May 2 nd ,2019. 4.R.C.Schnur et al.Preparation of N-phenylquinazoline-4-amines as neoplasm inhibitors.WO9630347,Oct.3 rd ,1996. 5.N.Blaquiere et al.,HETERO-BIFUNCTIONAL DEGRADER COMPOUNDS AND THEIR USE AS MODULATORS OF TARGETED UBIQUINATION(VHL)WO2019183523,Sept 26 th ,2019 6.Y.Jia et al.,Overcoming EGFR(T790M)and EGFR(C797S)resistance with mutant-selective allosteric inhibitors.Nature 534,129-132(2016). 7.G.Yu,L.G.Wang,Y.Han,Q.Y.He,clusterProfiler:an R package for comparing biological themes among gene clusters.OMICS 16,28(2012) 8.K.Street et al.,Slingshot:cell lineage and pseudotime inference for single-cell transcriptomics.BMC Genomics 19,477(2018). 9.P.B.Chapman et al.,Improved survival with vemurafenib in melanoma with BRAF V600E mutation.N Engl J Med 364,2507-2516(2011). 10.P.Lito,N.Rosen,D.B.Solit,Tumor adaptation and resistance to RAF inhibitors.Nat Med 19,1401-1409(2013). 11.Y.L.Choi et al.,EML4-ALK mutations in lung cancer that confer resistance to ALK inhibitors.N Engl J Med 363,1734-1739(2010). 12.M.E.Gorre et al.,Clinical resistance to STI-571 cancer therapy caused by BCR-ABL gene mutation or amplification.Science 293,876-880(2001). 13.S.Kobayashi et al.,EGFR mutation and resistance of non-small-cell lung cancer to gefitinib.N Engl J Med 352,786-792(2005). 14.W.Pao et al.,Acquired resistance of lung adenocarcinomas to gefitinib or erlotinib is associated with a second mutation in the EGFR kinase domain.PLoS Med 2,e73(2005). 15.R.L.Yauch et al.,Epithelial versus mesenchymal phenotype determines in vitro sensitivity and predicts clinical activity of erlotinib in lung cancer patients.Clin Cancer Res 11,8686-8698(2005). 16.A.H.Davies,H.Beltran,A.Zoubeidi,Cellular plasticity and the neuroendocrine phenotype in prostate cancer.Nat Rev Urol 15,271-286(2018). 17.H.Kitai et al.,Epithelial-to-Mesenchymal Transition Defines Feedback Activation of Receptor Tyrosine Kinase Signaling Induced by MEK Inhibition in KRAS-Mutant Lung Cancer.Cancer Discov 6,754-769(2016). 18.H.Uramoto et al.,Epithelial-mesenchymal transition in EGFR-TKI acquired resistant lung adenocarcinoma.Anticancer Res 30,2513-2517(2010). 19.G.M.Burslem,C.M.Crews,Proteolysis-Targeting Chimeras as Therapeutics and Tools for Biological Discovery.Cell 181,102-114(2020). 20.R.Verma,D.Mohl,R.J.Deshaies,Harnessing the Power of Proteolysis for Targeted Protein Inactivation.Mol Cell 77,446-460(2020). 21.C.M.Olson et al.,Pharmacological perturbation of CDK9 using selective CDK9 inhibition or degradation.Nat Chem Biol 14,163-170(2018). 22.I.You et al.,Discovery of an AKT Degrader with Prolonged Inhibition of Downstream Signaling.Cell Chem Biol 27,66-73 e67(2020). 23.A.Dixit et al.,Perturb-Seq:Dissecting Molecular Circuits with Scalable Single-Cell RNA Profiling of Pooled Genetic Screens.Cell 167,1853-1866 e1817(2016). 24.C.Weinreb,A.Rodriguez-Fraticelli,F.D.Camargo,A.M.Klein,Lineage tracing on transcriptional landscapes links state to fate during differentiation.Science 367,(2020). 25.T.J.Lynch et al.,Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib.N Engl J Med 350,2129-2139(2004). 26.J.G.Paez et al.,EGFR mutations in lung cancer:correlation with clinical response to gefitinib therapy.Science 304,1497-1500(2004). 27.K.Sankar,S.M.Gadgeel,A.Qin,Molecular therapeutic targets in non-small cell lung cancer.Expert Rev Anticancer Ther,(2020). 28.J.C.Soria et al.,Osimertinib in Untreated EGFR-Mutated Advanced Non-Small-Cell Lung Cancer.N Engl J Med 378,113-125(2018). 29.A.Leonetti et al.,Resistance mechanisms to osimertinib in EGFR-mutated non-small cell lung cancer.Br J Cancer 121,725-737(2019). 30.D.P.Bondeson et al.,Catalytic in vivo protein knockdown by small-molecule PROTACs.Nat Chem Biol 11,611-617(2015). 31.K.N.Shah et al.,Aurora kinase A drives the evolution of resistance to third-generation EGFR inhibitors in lung cancer.Nat Med 25,111-118(2019). 32.S.V.Sharma et al.,A chromatin-mediated reversible drug-tolerant state in cancer cell subpopulations.Cell 141,69-80(2010). 33.L.A.Byers et al.,An epithelial-mesenchymal transition gene signature predicts resistance to EGFR and PI3K inhibitors and identifies Axl as a therapeutic target for overcoming EGFR inhibitor resistance.Clin Cancer Res 19,279-290(2013). 34.N.Okura et al.,ONO-7475,a Novel AXL Inhibitor,Suppresses the Adaptive Resistance to Initial EGFR-TKI Treatment in EGFR-Mutated Non-Small Cell Lung Cancer.Clin Cancer Res 26,2244-2256(2020). 35.H.Taniguchi et al.,AXL confers intrinsic resistance to osimertinib and advances the emergence of tolerant cells.Nat Commun 10,259(2019). 36.Z.Zhang et al.,Activation of the AXL kinase causes resistance to EGFR-targeted therapy in lung cancer.Nat Genet 44,852-860(2012). 37.A.N.Hata et al.,Tumor cells can follow distinct evolutionary paths to become resistant to epidermal growth factor receptor inhibition.Nat Med 22,262-269(2016). 38.M.C.Wagle et al.,A transcriptional MAPK Pathway Activity Score(MPAS)is a clinically relevant biomarker in multiple cancer types.NPJ Precis Oncol 2,7(2018). 39.Y.Jia et al.,Overcoming EGFR(T790M)and EGFR(C797S)resistance with mutant-selective allosteric inhibitors.Nature 534,129-132(2016). 40.C.To et al.,Single and Dual Targeting of Mutant EGFR with an Allosteric Inhibitor.Cancer Discov 9,926-943(2019). 41.S.J.Marciniak et al.,CHOP induces death by promoting protein synthesis and oxidation in the stressed endoplasmic reticulum.Genes Dev 18,3066-3077(2004). 42.P.Walter,D.Ron,The unfolded protein response:from stress pathway to homeostatic regulation.Science 334,1081-1086(2011). 43.E.Timosenko et al.,Nutritional Stress Induced by Tryptophan-Degrading Enzymes Results in ATF4-Dependent Reprogramming of the Amino Acid Transporter Profile in Tumor Cells.Cancer Res 76,6193-6204(2016). 44.H.H.Rabouw et al.,Small molecule ISRIB suppresses the integrated stress response within a defined window of activation.Proc Natl Acad Sci U S A 116,2097-2102(2019). 45.D.Buckley,et al.Targeting the von Hippel-Lindau E3 ubiquitin ligase using small molecules to disrupt the VHL / HIF-1α interaction.J Am Chem Soc 134,4465-8(2012). 46.C.Hetz,K.Zhang,R.J.Kaufman,Mechanisms,regulation and functions of the unfolded protein resonse.Nat Rev Mol Cell Biol 21,421-438(2020). 47.S.Stockwell,et al.,Mechanism-based screen for G1-S checkpoint acitvators identifies a selective activator of EIFAK3 / PERK signaling.PLoS One 7,e28568(2012). 48.J.Bruch,et al.,PERK activation mitigates tau pathology in vitro and in vivo.EMBO Mol Med 9,371-384(2017).

Claims

1. A method for screening cells for a trait, comprising: (a) performing a first sequencing of RNA and / or DNA on a subset of a plurality of barcoded cells, wherein each barcoded cell comprises a single unique barcode; (b) culturing the plurality of barcoded cells for a first period of time in the presence of a selection pressure, thereby forming a first plurality of cells; (c) performing a second sequencing of RNA and / or DNA on a subset of the first plurality of cells; (d) culturing the first plurality of cells for a second period of time in the presence of the selection pressure, thereby forming a second plurality of cells; (e) performing a third sequencing of RNA and / or DNA on at least a subset of the second plurality of cells; and (f) determining the level of the barcode sequenced in the first sequencing, the second sequencing, and / or the third sequencing; A method comprising:

2. A method for screening cells for a response trait to a therapeutic agent, comprising: (a) performing a first sequencing of RNA and / or DNA on a subset of a plurality of barcoded cells, wherein each barcoded cell comprises a single unique barcode; (b) culturing the plurality of barcoded cells for a first period of time in the presence of the therapeutic agent, thereby forming a first plurality of cells; (c) performing a second sequencing of RNA and / or DNA on a subset of the first plurality of cells; (d) culturing the first plurality of cells with the therapeutic agent for a second period of time, thereby forming a second plurality of cells; (e) performing a third sequencing of RNA and / or DNA on a subset of the second plurality of cells; and (f) determining the level of the barcode sequenced in the first sequencing, the second sequencing, and / or the third sequencing; A method comprising:

3. The method of claim 1 or 2, wherein the plurality of barcoded cells are grown in culture prior to step (a) or step (b).

4. A method for comparing responses to a selection pressure, comprising: (a) performing a first sequencing of RNA and / or DNA from a first plurality of barcoded cells and / or a second plurality of barcoded cells substantially similar to the first plurality of barcoded cells, wherein each barcoded cell of the first plurality of barcoded cells comprises a single unique barcode; (b) culturing the first plurality of barcoded cells in the presence of a first selection pressure, thereby forming a first plurality of cells; (c) culturing the second plurality of barcoded cells in the presence of a second selection pressure, thereby forming a second plurality of cells; (d) performing a second sequencing of RNA and / or DNA from the first plurality of cells and / or the second plurality of cells; (e) culturing said first plurality of cells in the presence of said first selection pressure, thereby forming a third plurality of cells; (f) culturing said second plurality of cells in the presence of said second selection pressure, thereby forming a fourth plurality of cells; (g) performing a third sequencing of RNA and / or DNA from the third plurality of cells and / or the fourth plurality of cells; and (h) determining the level of one or more barcodes sequenced in said first sequencing, second sequencing and / or third sequencing. Including, Steps (e)-(g) are repeated one or more iterations, thereby forming one or more subsequent plurality of cells and one or more subsequent sequencing. method.

5. The method of claim 4, wherein step (h) further comprises determining the level of one or more barcodes in the one or more subsequent sequencing steps.

6. The method of claim 4 or 5, wherein the first plurality of barcoded cells and the second plurality of barcoded cells comprise multiple clonal populations, and each cell within a single clonal population comprises an identical single unique barcode.

7. A method according to any one of claims 3 to 6, wherein the single unique barcode is a unique combination of barcodes.

8. The method of claim 6, wherein the relative abundance of cells in each clonal population is approximately equal to the number of cells in each other clonal population in the first plurality of barcoded cells and the second plurality of barcoded cells.

9. The method of claim 8, wherein from the first sequencing step, the relative abundance of cells in a clonal population is determined relative to the number of cells containing the barcode(s).

10. The method of any one of claims 3 to 9, further comprising identifying a barcode(s) that is concentrated in the first plurality of cells and / or the second plurality of cells.

11. The method of claim 10, further comprising identifying one or more genes having higher expression levels in cells containing the enriched barcode(s).

12. The method of any one of claims 3 to 11, further comprising determining a first expression level of a gene in cells having the barcode(s) enriched in the first plurality of cells and / or the third plurality of cells and / or the one or more subsequent plurality of cells based on the second sequencing and / or the third sequencing and / or the one or more subsequent sequencings, and a second expression level of the gene in the second plurality of barcoded cells and / or the fourth plurality of cells and / or the one or more subsequent plurality of cells based on the first sequencing, the second sequencing and / or the third sequencing and / or the one or more subsequent sequencings.

13. The method of claim 12, further comprising comparing the first expression level of the gene with the second expression level of the gene.

14. A method for screening cells for a trait in the cells, comprising: (a) culturing a mixture of cells in the presence of a first selection pressure for a first period of time and obtaining, at the end of said first period of time, second genetic, transcriptomic and / or proteomic information for at least one member of a surviving clonal population from within said mixture of cells, said mixture of cells comprising a plurality of clonal populations, each clonal population comprising an identifier unique to the respective clonal population, and initial genetic, transcriptomic and / or proteomic information of at least one representative member of each clonal population being known; (b) subjecting the mixture of cells subjected to the first selection pressure to a second selection pressure for a second period of time, and obtaining, at the end of the second period of time, a third genetic, transcriptomic and / or proteomic information of at least one member of a surviving clonal population from within the mixture of cells; and (c) determining the level of the clonal population present in the final mixture of cells based on the unique identifier for the clonal population. A method comprising:

15. The method of claim 14, wherein step (a) or (b) is repeated one or more times.

16. The method of claim 14, wherein steps (a) and (b) are repeated one or more times.

17. The method of any one of claims 14 to 16, further comprising identifying an adaptive trait, said adaptive trait being a genetic and / or proteomic trait present or absent in a clonal population in said final mixture of cells.

18. The method of claim 17, wherein the adaptive trait is the presence or absence of a gene, allele, genetic modification, transcript or protein; or a change in a gene, allele, transcript or protein when comparing the first, second and / or third and / or one or more subsequent genetic, transcriptomic and / or proteomic information obtained.

19. A method according to any one of claims 14 to 18, wherein step (a) comprises obtaining a fourth genetic information, transcriptomic information and / or proteomic information of at least one member of a second surviving clonal population from within the mixture of cells.

20. A method according to any one of claims 14 to 19, wherein step (b) comprises obtaining a fifth genetic information, transcriptomic information and / or proteomic information of at least one member of a second surviving clonal population from within the mixture of cells.

21. The method of any one of claims 14 to 20, further comprising comparing information from the first genetic information, transcriptomic information, and / or proteomic information, the second genetic information, transcriptomic information, and / or proteomic information, and / or the third genetic information, transcriptomic information, and / or proteomic information, and / or one or more subsequent genetic information, transcriptomic information, and / or proteomic information.

22. A method according to any one of claims 14 to 21, wherein obtaining the genetic information, transcriptomic information, and / or proteomic information comprises RNA-seq, single-cell RNA-seq, DNA sequencing, epigenetic sequencing or protein sequencing.

23. The method of any one of claims 14 to 22, wherein the first selective pressure and / or the second selective pressure comprises treatment with a therapeutic agent, contact with a contaminant, genome engineering, engraftment into a host, culture conditions, growth conditions, contact with a stimulus, or contact with other cells.

24. A method for identifying a cellular program that facilitates adaptation to pressure, comprising the steps of: (a) transducing cells with multiple barcodes such that each cell contains a single unique barcode; (b) expanding the cells in culture to generate a clonal starting cell pool of cells containing each barcode; (c) obtaining a first genetic information, transcriptomic information, and / or proteomic information from a first subset of the starting cell pool; (d) culturing a second subset of the starting cell pool in the presence of selective pressure to expand the starting cell pool and form an intermediate cell pool; (e) obtaining second genetic information, transcriptomic information, and / or proteomic information from a first subset of the intermediate cell pool; (f) continuing to culture a second subset of said intermediate cell pool in the presence of said selective pressure to expand said intermediate cell pool and form a final cell pool; (g) obtaining a third genetic information, transcriptomic information, and / or proteomic information from at least a subset of said final cell pool; (h) quantifying the level of each barcode in the final cell pool, the intermediate cell pool, and / or the starting cell pool; (i) assigning cells with barcodes that are enriched in the final cell pool as winning clones, and / or assigning cells with barcodes that are depleted in the final cell pool as losing clones; and (j) determining genetic mutations, transcriptional programs, and / or protein expression associated with at least one winning clone and / or at least one losing clone; Including, A method in which an approximately equal number of clones of cells containing each barcode are used to create the starting cell pool.

25. The method of claim 24, wherein the number of cell clones is normalized to the number of cells containing each barcode obtained in step (c).

26. The method of claim 24 or 25, wherein steps (d) and (e) are repeated one or more iterations, thereby obtaining one or more additional intermediate cell pools and one or more additional intermediate genetic information, transcriptomic information, and / or proteomic information.

27. ​​The method described in claim 26, wherein step (h) further comprises quantifying the level of each barcode in the one or more additional intermediate cell pools.

28. The method of any one of claims 24 to 27, wherein the single unique barcode is a unique combination of barcodes.

29. The method of any one of claims 24 to 27, wherein the selective pressure comprises treatment with a therapeutic agent, contact with a contaminant, genome engineering, engraftment into a host, culture conditions, growth conditions, contact with a stimuli, or contact with other cells.

30. The method of any one of claims 24 to 29, wherein obtaining the genetic information, transcriptomic information, and / or proteomic information comprises RNA-seq, single-cell RNA-seq, DNA sequencing, epigenetic sequencing or protein sequencing.