human genome wide crispri v2 libraries  (Addgene inc)

Journal: bioRxiv

Article Title: Systematic identification of chromatin organizers as tuners of intratumoral heterogeneity

doi: 10.64898/2026.04.18.719392

Figure Lengend Snippet: (A) Schematic of the Perturb-seq workflow. MDA-MB-231 cells were transduced with CRISPRi or CRISPRa sgRNAs targeting the 16 in vivo -validated COs, followed by 10X 3 ′ scRNA-seq. Genome-wide CV was computed for each perturbation and compared to size-matched random control groupings. (B) Mean and standard error of mean barplots for genome-wide CV change upon CRISPRi (left) or CRISPRa (right) perturbation of RNF8 and MIS18A. A value of 1 indicates no change relative to non-targeting control. P -values by one-sample two-tailed Student’s t -test ( μ = 1). (C, D) Per-gene transcript CV in CRISPRi cells versus non-targeting controls for RNF8 (C) and MIS18A (D) . Each horizontal line represents a gene; P -values by two-sided paired Student’s t -test. Percentages indicate proportion of genes that decrease (top) or increase (bottom) in CV relative to non-targeting control. (E) CV analysis in an independent patient tumor scRNA-seq cohort . Patients were stratified into quartiles by RNF8 or MIS18A expression ( n = 4 per quartile). P -values by two-sided Wilcoxon signed-rank test.

Article Snippet: For our computationally nominated CRISPRi library, a library consisting of guides targeting 185 elements (5 sgRNAs each for 37 genes) was designed and ordered from Twist Biosciences.

Techniques: Transduction, In Vivo, Genome Wide, Control, Two Tailed Test, Expressing

Journal: bioRxiv

Article Title: Systematic identification of chromatin organizers as tuners of intratumoral heterogeneity

doi: 10.64898/2026.04.18.719392

Figure Lengend Snippet: (A, B) Mean and standard error of mean (SEM) barplots for genome-wide CV change upon CRISPRi (A) or CRISPRa (B) perturbation of all 16 in vivo -validated COs. A value of 1 indicates no change relative to non-targeting control. Size-matched random cell groupings serve as empirical null. P -values by one-sample two-tailed Student’s t -test. (C) Per-sgRNA CV change for each guide in the CRISPRi and CRISPRa libraries. Middle and right bars show the two individual sgRNAs per gene. P -values by two-tailed Student’s t -test, adjusted with the Bonferroni procedure. (D, E) Per-gene transcript CV in CRISPRa cells versus non-targeting controls for RNF8 (D) and MIS18A (E) . Each horizontal line represents a gene; P -values by two-sided paired Student’s t -test. Percentages indicate proportion of genes that decrease (top) or increase (bottom) in CV relative to non-targeting control. (F, G) Cell-cycle phase distribution of cells with the indicated sgRNA perturbation in CRISPRi (F) and CRISPRa (G) lines. Neither RNF8 nor MIS18A perturbation significantly altered cell-cycle composition. P -values by two-sided Wilcoxon signed-rank test. (H, I) Mean of the genome-wide gene expression means for the indicated sgRNA perturbations in CRISPRi (H) and CRISPRa (I) with standard error of mean (SEM) shown. P -values by one-sample two-tailed Student’s t -test, adjusted with the Benjamini–Hochberg procedure.

Article Snippet: For our computationally nominated CRISPRi library, a library consisting of guides targeting 185 elements (5 sgRNAs each for 37 genes) was designed and ordered from Twist Biosciences.

Techniques: Genome Wide, In Vivo, Control, Two Tailed Test, Gene Expression

Journal: bioRxiv

Article Title: Systematic identification of chromatin organizers as tuners of intratumoral heterogeneity

doi: 10.64898/2026.04.18.719392

Figure Lengend Snippet: (A) Immunoblot of CENP-A in MDA-MB-231 CRISPRi/a lines. Top: CENP-A normalized to tubulin loading control. Bottom: quantification of tubulin-normalized CENP-A signal relative to non-targeting con-trol. (B) RT-qPCR of minor satellite ncRNAs in MDA-MB-231 CRISPRi/a lines at 48 and 96 hours post-seeding ( n = 4). Values are GAPDH-normalized. No significant accumulation of minor satellite ncRNAs was observed upon MIS18A knockdown, indicating that centromeric histone modifications remain intact. (C) CENP-A im-munofluorescence in MDA-MB-231 CRISPRi/a lines. Top: representative images of CENP-A and DAPI signal. Bottom: quantification of CENP-A as a fraction of DAPI-identified clusters. MIS18A knockdown did not re-duce CENP-A signal. Scale bar, 50 μ m. (D) Cell-painting morphology heterogeneity scores for MDA-MB-231 CRISPRa lines with the indicated sgRNAs, with or without the DNMT3B inhibitor nanaomycin. Nanaomycin rescued MIS18A-overexpression-dependent morphological changes, indicating that MIS18A affects cellular phenotype through DNMT3B-mediated methylation. P -values by one-sided Wilcoxon signed-rank test.

Article Snippet: For our computationally nominated CRISPRi library, a library consisting of guides targeting 185 elements (5 sgRNAs each for 37 genes) was designed and ordered from Twist Biosciences.

Techniques: Western Blot, Control, Quantitative RT-PCR, Knockdown, Over Expression, Methylation

Journal: bioRxiv

Article Title: A genome-wide CRISPR screen reveals cancer-specific regulators of hyaluronan binding and cellular invasion

doi: 10.64898/2026.02.06.704476

Figure Lengend Snippet: A) Chemical structure of Fluoresceinamine-Hyaluronan (FL-HA). B ) MDA-MB-231 cells were incubated with FL-HA for 1 hour on ice at a 1:100 dilution and subsequently analyzed by flow cytometry. In some conditions, cells were pre-treated with a CD44 blocking antibody (HERMES-1 clone). For every 10,000 cells, 1 µL of blocking antibody was added to cells 30 minutes prior to incubation with FL-HA. C ) MDA-MB-231-dCas9KRAB cells were transduced with an sgRNA against CD44. An antibody against CD44 (Hermes-3, 1:100) was then pre-complexed with an anti-mouse Alexa 488 secondary antibody (1.5 µg/mL) for 30 minutes and then incubated with the cells on ice. Cells were then analyzed by flow cytometry. A representative plot is shown. D) The median fluorescence intensity of CD44 staining is plotted for n=3 independent replicates. Statistical significance was determined using a Student’s two-tailed t-test, * indicates p<0.05, *** indicates p<0.001. E) MDA-MB-231-dCas9KRAB cells were transduced with an sgRNA against CD44 as in 1C . Cells were then stained with FL-HA as in 1B . A representative plot is shown. F) The median fluorescence intensity of FL-HA staining is plotted for n=3 independent replicates. Statistical significance was determined using a Student’s two-tailed t-test, * indicates p<0.05. G) Workflow of genome-wide CRISPRi screen. MDA-MB-231-dCas9KRAB cells were transduced with an sgRNA library containing 104,000 sgRNAs. Cells were subsequently stained with FL-HA as in 1B and sorted by FACS to isolate low-staining (bottom 20%) and high-staining (top 20%) populations. sgRNAs were sequenced using an Illumina platform and sgRNA enrichment scores were calculated using MAGeCK.

Article Snippet: MDA-dCas9–KRAB cells were transduced with a genome-wide CRISPRi library of sgRNAs consisting of 5 sgRNAs/gene (AddGene, Cat # 83969) as described above.

Techniques: Incubation, Flow Cytometry, Blocking Assay, Transduction, Fluorescence, Staining, Two Tailed Test, Genome Wide

Journal: bioRxiv

Article Title: Ribosome-associated quality control of aberrant protein production during amino acid limitation

doi: 10.64898/2026.01.14.699605

Figure Lengend Snippet: (A) A dual reporter system with a Flag-mCherry CGG -4xCGG-DHFR CGG (internal control, no translation disruption) and a Flag-YFP CGG -4xAGA-DHFR CGG was integrated into the AAVS1 locus in cells expressing CRISPRi or CRISPRa machinery (see Methods) to query the effect of a genome-wide library of CRISPRi/a sgRNAs on specific translation disruption signal. Flow cytometry was used to sort high and low responding populations. (B-D) Total score (phenotype score * Mann-Whitney (M-W) p-value, see Methods) versus hit rank for highest (high YFP) and lowest (low YFP) scoring genes in K562 CRISPRi (B), 293T CRISPRa (C), and 293T CRISPRi (D) screens. Hits are labeled and colored by associated function. Dashed line marks score cut-off where FDR = 0.25. (E) Diagram connecting major screen hit pathways. Hits are colored by direction of phenotype; blue gene labels reduce translation disruption product accumulation, red gene labels increase translation disruption product accumulation.

Article Snippet: Genome-wide CRISPRi and CRISPRa sgRNA libraries (hCRISPRi_v2: Addgene #83969 and #83970; hCRISPRa_v2: #83978 and #83979) were amplified, packaged into lentiviral particles, and titers were determined as described in ref . For K562, CRISPRi parental cells (187 million) were infected at a multiplicity of infection (MOI) of 0.28 and selected using 2 μg/mL puromycin (Sigma) for 6 days starting 48 h post-transduction.

Techniques: Control, Disruption, Expressing, Genome Wide, Flow Cytometry, MANN-WHITNEY, Labeling

Journal: bioRxiv

Article Title: Ribosome-associated quality control of aberrant protein production during amino acid limitation

doi: 10.64898/2026.01.14.699605

Figure Lengend Snippet: (A) Schematic outlining how the screen with reporters and sorting scheme depicted in was performed. Cells were arginine limited for 3 days, sorted, recovered, and re-sorted into the same bin after a second period of arginine limitation. After recovery, guide RNAs were sequenced to calculate enrichment scores. (B,C) Western blot (B) and flow cytometry (C) validation of selected hits with negative or positive phenotype scores across various pathways in K562 cells. Cells expressing CRISPRi targeting hits (or non-targeting controls; NTC) and dual color translation disruption reporters (Flag-YFP CGG -4xAGA-DHFR CGG and Flag-mCherry CGG -4xCGG-DHFR CGG ) were arginine limited for 3 days to assess translation disruption product levels (B,C) and signaling responses through mTOR, GCN2 and ZAKα (B). In (B), “*” marks non-specific band from blot stripping and reprobing. (D) Western blot to assess translation disruption and GCN2 response in 293T cells overexpressing GADD34 or an NTC by CRISPRa (scFv-sfGFP-GCN4-VP64) and dual color translation disruption reporters (Flag-YFP CGG -4xAGA-DHFR CGG and Flag-mCherry CGG -4xCGG-DHFR CGG ), with or without limitation for arginine for 5 days and treatment with 40 nM ISRIB. (E) Western blot to assess translation disruption with or without limitation for leucine or arginine for 7 days and treatment with 250 nM Torin1 in MiaPaCa cells expressing the Flag-YFP CGG -2xAGA-DHFR CGG reporter. (F) Flow cytometry to assess translation disruption product accumulation upon limitation for arginine with or without GCN2 knockdown by CRISPRi and 250 nM Torin1 treatment in K562 cells expressing the dual color translation disruption reporters (Flag-YFP CGG -4xAGA-DHFR CGG and Flag-mCherry CGG -4xCGG-DHFR CGG ). (G-H) Western blots to assess phospho-p38 and -JNK response to arginine or leucine limitation and 0.1 μg/mL anisomycin treatment with or without treatment with 5 μM SB202190 (p38 inhibitor) in wild-type and GCN2 KO 293T cells. (I) Western blot to confirm ZAKα KO in wildtype and GCN2 KO 293T cells as indicated. (J-K) Western blots to assess phospho-p38 and -JNK response to 0.1 μg/mL anisomycin treatment or UV irradiation (J) or arginine or leucine limitation (K) in wild-type and GCN2 KO, ZAKα KO, and ZAKα+GCN2 double KO 293T cells as indicated. (L) Flow cytometry to assess reporter fluorescence upon limitation for arginine for 6 days with or without treatment with 5 μM SB202190 in wild-type and GCN2 KO 293T cells expressing the Flag-YFP CGG -4xAGA-DHFR CGG (“AGA”) or Flag-YFP CGG -4xCGG-DHFR CGG (“CGG”) reporter as indicated. (M) Flow cytometry to assess reporter fluorescence over time upon limitation for arginine with or without treatment with 5 μM SB202190 in K562 cells with (sgGCN2) or without (sgNTC) GCN2 CRISPRi knockdown expressing the dual color translation disruption reporters (Flag-YFP CGG -4xAGA-DHFR CGG and Flag-mCherry CGG -4xCGG-DHFR CGG ). (N) Change in translation disruption reporter (Flag-YFP CGG -4xAGA-DHFR CGG (“AGA”) or Flag-YFP CGG -4xAGA-DHFR CGG (“CGG”)) mRNA level upon arginine limitation for 3 days with or without treatment with 5 μM SB202190 in 293T cells. (O) Change in translation disruption reporter mRNA level (Flag-YFP CGG -4xAGA-DHFR CGG ) upon arginine limitation for 3 days with (sgLAMTOR2) or without (sgNTC) CRISPRi knockdown of LAMTOR2 in 293T cells expressing the dual color translation disruption reporters (Flag-YFP CGG -4xAGA-DHFR CGG and Flag-mCherry CGG -4xCGG-DHFR CGG ), relative to cells expressing control guide (sgNTC 1). (C,L,N,O) Error bars represent standard error of the mean of 3 replicates. (B,D,E, G-K) Full-length reporter product is indicated (α-GFP antibody used to detect YFP, α-vinc = α-vinculin).

Article Snippet: Genome-wide CRISPRi and CRISPRa sgRNA libraries (hCRISPRi_v2: Addgene #83969 and #83970; hCRISPRa_v2: #83978 and #83979) were amplified, packaged into lentiviral particles, and titers were determined as described in ref . For K562, CRISPRi parental cells (187 million) were infected at a multiplicity of infection (MOI) of 0.28 and selected using 2 μg/mL puromycin (Sigma) for 6 days starting 48 h post-transduction.

Techniques: Western Blot, Flow Cytometry, Biomarker Discovery, Expressing, Disruption, Stripping, Knockdown, Irradiation, Fluorescence, Control

Journal: bioRxiv

Article Title: Mapping kinase-dependent tumor immune adaptation with multiplexed single-cell CRISPR screens

doi: 10.64898/2026.01.08.698516

Figure Lengend Snippet: (a) Experimental schematic of CRISPRi/a perturbations in NY-ESO-1/HLA-A*02:01-engineered GBM cells co-cultured with T cells. (b) UMAP embeddings of cancer cells under CRISPRi (left) or CRISPRa (right) showing transcriptomic trajectories across E:T ratios. (c) Module-level changes in MHC-I, MHC-II, IFN-γ, and NF-κB pathway programs under CRISPRi versus CRISPRa. Colors indicate effect size, quantified as the β coefficient for the genotype term (perturbation) in a regression model evaluated under the no–T cell condition. (d) Schematic for MrVI analysis, and (e) Forest plots summarizing perturbation effects on T cell-induced program across E:T ratios from MrVI analysis. Color and circle size denote p values, while arrow length and thickness represent distance differences. The statistical tests were performed with a one-sided Mann-Whitney U test (Bonferroni-corrected, α < 0.05).

Article Snippet: Protospacer sequences targeting all genes perturbed in this study were sourced from the genome-wide human CRISPRi and CRISPRa v2 libraries developed by Horlbeck et al. (CROP-seq-OPTI, Addgene, Plasmid #106280, engineered to include a CRISPRi-optimized gRNA backbone) were synthesized by Integrated DNA Technologies (IDT).

Techniques: Cell Culture, MANN-WHITNEY

Journal: bioRxiv

Article Title: Mapping kinase-dependent tumor immune adaptation with multiplexed single-cell CRISPR screens

doi: 10.64898/2026.01.08.698516

Figure Lengend Snippet: (a) Decipher analysis of NTC cells from CRISPRa data reveals trajectories driven by increasing T cell concentration, shown in the learned 2D Decipher space with cells colored by E:T ratios (left) and corresponding Decipher time (pseudotime; right). (b) Genes significantly contribute to the Decipher trajectory, with the positions of IDO1 and SOD2 highlighted. (c) Single-cell trajectories of representative genes inferred by Decipher reveal E:T ratio-dependent transcriptomic responses. (d) Decipher-aligned CRISPRi and CRISPRa cells (left) projected into a shared 2D Decipher space together with NTC cells (right). (e) Decipher trajectories learned as a function of E:T ratio (left) and pseudotime (right) for CRISPRi and CRISPRa conditions. (f) Comparison of CRISPRi and CRISPRa perturbations highlights conserved effects on immune checkpoint expression ( CD274 ) and tumor-intrinsic T cell-responsive genes ( CSF3 ). (g) Density plots illustrate distribution shifts of cancer cells along Decipher time following perturbation of selected kinases ( EPHA2 , STK40 ), indicating that kinase disruption alters the progression of cancer cell states under T cell pressure (p = 0.033639216, p = 0.008503096 with Kolmogorov-Smirnov test).

Article Snippet: Protospacer sequences targeting all genes perturbed in this study were sourced from the genome-wide human CRISPRi and CRISPRa v2 libraries developed by Horlbeck et al. (CROP-seq-OPTI, Addgene, Plasmid #106280, engineered to include a CRISPRi-optimized gRNA backbone) were synthesized by Integrated DNA Technologies (IDT).

Techniques: Concentration Assay, Comparison, Expressing, Disruption

Journal: bioRxiv

Article Title: Genome-scale perturb-seq in primary human CD4+ T cells maps context-specific regulators of T cell programs and human immune traits

doi: 10.64898/2025.12.23.696273

Figure Lengend Snippet: ( A ) Experimental design of genome-scale perturb-seq in CD4+ T cells from multiple human donors across conditions (see Methods). ( B ) Number of cells (y-axis) for each biological sample (x-axis, condition-donor) with transcriptome passing quality control. Bars are colored by the outcome of guide RNA assignment (NTC: non-targeting controls). ( C ) Distribution of the number of cells harboring each gene perturbation (x-axis, log10 scale) across culture conditions (rows). The dotted line denotes the median. ( D ) Fraction of guides inducing significant on-target knock-down (y-axis) versus baseline expression of perturbed genes (x-axis). Each point represents 100 genes grouped by similar expression levels. Dotted vertical lines denote mean expression (log-normalized counts) per bin. ( E ) Comparison of trans effects between perturb-seq and arrayed knockout (KO) RNA-seq [ , ]: for each perturbed gene (x-axis), Pearson correlation of gene expression Log fold changes (LFCs) across all measured genes (y-axis) is shown. Blue bars: comparison of same gene perturbed in both platforms; grey bars: comparison on different perturbed genes. Error bars: 95% confidence interval (CI) from bootstrapping. Dotted line: theoretical maximum correlation given LFC noise. ( F ) Comparison of trans effects between perturb-seq and published CRISPRi FACS-based screens [ , , ]. For each measured gene (x-axis), Pearson correlation of expression LFCs across all perturbed genes (y-axis) is shown. Colored bars: same gene measured in both platforms; grey-scale bars: expression-matched different genes. Error bars: 95% CI from bootstrapping. Stimulation conditions of FACS screens are indicated below x-axis. ( G ) Pearson correlation between LFCs from downsampled and held-out validation data (5 lanes, all donors) for genes significant at 10% FDR. X-axis: fraction of perturbed cells (20–100%, ~55–450 cells). Colors: number of donors (1–4). Error bars: variability across independent downsampling splits and perturbed genes. Aggregated results across 12 perturbed genes are shown (see Suppl. Figure 9 for per-gene results). Grey dotted line: maximum correlation between perturbation effects in independent held-out splits.

Article Snippet: Our genome-scale CRISPRi library (to be available on Addgene) was cloned into PJY390 by Golden Gate Cloning as described previously [ ].

Techniques: Control, Knockdown, Expressing, Comparison, Knock-Out, RNA Sequencing, Gene Expression, Biomarker Discovery