[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7CV2FVB7RGJVRRZYH8NQHED":3,"skill-same:01M7CV2FVB7RGJVRRZYH8NQHED":84},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":52,"commitSha":53,"skillDirectory":54,"forksCount":55,"repositoryHtmlUrl":56,"repositoryHomepageUrl":6,"ownerHtmlUrl":57,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":58,"install":59,"resources":71},"01M7CV2FVB7RGJVRRZYH8NQHED","bio-workflows-crispr-screen-pipeline",null,"End-to-end pooled and single-cell CRISPR screen analysis from FASTQ to hit genes. Orchestrates library design QC, guide counting, six-stage screen QC (plasmid Gini, replicate Pearson, CEGv2 PR-AUC, copy-number artifact), method-appropriate hit calling across MAGeCK RRA\u002FMLE, BAGEL2, drugZ, JACKS, and Chronos, cancer-cell-line copy-number correction (CRISPRcleanR \u002F Chronos), batch correction for multi-batch screens, and the specialized branches for combinatorial paralog screens, single-cell Perturb-seq, base-editor variant-function screens, prime-editor screens, and in vivo bottleneck-aware screens. Use when analyzing any pooled CRISPR screen end-to-end, matching the hit-calling method to the experimental design, integrating copy-number correction into the pipeline, or branching the workflow for single-cell, combinatorial, base-editor, prime-editor, or in vivo variants.","mrsonord2240\u002Fopenscience-specialists","mrsonord2240","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F76487430?v=4","01M7CTEM40S43GSJZ9A3YCCXZK","Samuel Nord","01M7CTEM4K69JD1GBYNHVJ9ZP4",2,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fmrsonord2240\u002Fopenscience-specialists\u002Fblob\u002Fmain\u002Fspecialists\u002Fcrispr-screen-analyst\u002Fversions\u002F1.0.0\u002Fpackage\u002Fskills\u002Fbio-workflows-crispr-screen-pipeline\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: bio-workflows-crispr-screen-pipeline\ndescription: End-to-end pooled and single-cell CRISPR screen analysis from FASTQ to hit genes. Orchestrates library design QC, guide counting, six-stage screen QC (plasmid Gini, replicate Pearson, CEGv2 PR-AUC, copy-number artifact), method-appropriate hit calling across MAGeCK RRA\u002FMLE, BAGEL2, drugZ, JACKS, and Chronos, cancer-cell-line copy-number correction (CRISPRcleanR \u002F Chronos), batch correction for multi-batch screens, and the specialized branches for combinatorial paralog screens, single-cell Perturb-seq, base-editor variant-function screens, prime-editor screens, and in vivo bottleneck-aware screens. Use when analyzing any pooled CRISPR screen end-to-end, matching the hit-calling method to the experimental design, integrating copy-number correction into the pipeline, or branching the workflow for single-cell, combinatorial, base-editor, prime-editor, or in vivo variants.\ntool_type: mixed\nprimary_tool: MAGeCK\nworkflow: true\ndepends_on:\n  - crispr-screens\u002Flibrary-design\n  - crispr-screens\u002Fscreen-qc\n  - crispr-screens\u002Fmageck-analysis\n  - crispr-screens\u002Fbagel-essentiality\n  - crispr-screens\u002Fdrugz-chemogenomic\n  - crispr-screens\u002Fjacks-analysis\n  - crispr-screens\u002Fhit-calling\n  - crispr-screens\u002Fcopy-number-correction\n  - crispr-screens\u002Fbatch-correction\n  - crispr-screens\u002Fcrispresso-editing\n  - crispr-screens\u002Fbase-editing-analysis\n  - crispr-screens\u002Fprime-editing-screens\n  - crispr-screens\u002Fperturb-seq-analysis\n  - crispr-screens\u002Fcombinatorial-screens\n  - crispr-screens\u002Fin-vivo-screens\nqc_checkpoints:\n  - after_counting: \">65% mapping rate; \u003C0.5% zero-count in plasmid; Gini \u003C0.1 on plasmid\"\n  - after_qc: \"Replicate Pearson on log-counts >=0.8 (MAGeCK-VISPR floor; >0.85 acceptable, >0.95 ideal); Spearman >0.7; CEGv2 PR-AUC >0.7\"\n  - after_cn_correction: \"Spearman ρ between CN and gene LFC abs \u003C0.10 post-correction (literature 'significant bias' band; \u003C0.05 is a stricter target). Requires a matched CN profile, which the unsupervised CRISPRcleanR path never loads -- compute in crispr-screens\u002Fcopy-number-correction, or use Chronos, which takes CN as input\"\n  - after_hit_calling: \"Tier-1 hits = 3-method consensus; Tier-2 = 2 of 3; Tier-3 = single-method exploratory\"\n---\n\n## Version Compatibility\n\nReference examples tested with: MAGeCK 0.5.9+, BAGEL2 1.0.5+, drugZ Aug 2019+, JACKS 0.2.0+, Chronos 2.0+, CRISPRcleanR 3.0+ (R), Pertpy 0.6+, PRIDICT2, CRISPResso2 2.2.14+, MAGeCKFlute 2.0+, pandas 2.2+, numpy 1.26+, matplotlib 3.8+.\n\nBefore using code patterns, verify installed versions match. If versions differ:\n- CLI: `mageck --version`, `BAGEL.py fc --help`, `drugz -h`, `CRISPResso --version`\n- Python: `pip show pertpy scanpy anndata` (mageck-vispr via conda `mageck --version`; JACKS\u002FChronos are GitHub installs — check their repos)\n- R: `packageVersion('CRISPRcleanR')`, `packageVersion('MAGeCKFlute')`\n\nIf code throws ImportError, AttributeError, or TypeError, introspect the installed package and adapt the example to match the actual API rather than retrying.\n\n## CRISPR Screen Pipeline\n\n**\"Analyze my pooled or single-cell CRISPR screen end-to-end\"** -> Pick the screen design branch, run guide counting, audit six QC stages, apply copy-number and batch correction as needed, run the design-matched hit-calling method, and consolidate across methods for high-confidence hits.\n\nThis is a workflow skill: it owns the chaining decisions and hand-offs, not the internals of any one step.\n\n## The governing principle\n\nEvery LFC, QC gate, and hit call is computed against a reference that is committed once at library-order time; a wrong-but-silent commitment invalidates the endpoint with no error thrown.\n\n1. **The guide LIBRARY definition (guide->gene map + control classes) is the denominator, the calibrator, and the training reference — committed once.** The library must carry non-targeting controls (NTCs, ~1%, the null distribution) AND CEGv2 reference essentials + NEGv1 non-essentials (the positive\u002Fnegative calibrators for PR-AUC and BAGEL2\u002FChronos priors). NTCs calibrate the null\u002FFDR; CEGv2\u002FNEGv1 calibrate PR-AUC — swapping or dropping a class silently breaks FDR or QC.\n2. **The baseline choice has a right answer and rescales every hit.** Dropout\u002Fenrichment LFC is against a baseline: plasmid pool for the cloning-bottleneck baseline, Day-0\u002FT0 for the biology baseline, and vehicle (NOT Day-0) for drug screens — drug-vs-Day-0 conflates drug effect with normal proliferation.\n3. **Copy-number correction MUST precede hit calling in cancer cell lines.** Multiple simultaneous Cas9 cuts at amplified loci trigger a gene-independent DNA-damage\u002FG2 arrest (Aguirre 2016; Munoz 2016; the effect appears in both TP53-mutant and TP53-wild-type lines, though Aguirre 2016 found TP53 status correlates with its magnitude -- separately, Ihry 2018 \u002F Haapaniemi 2018 report p53-dependent toxicity of Cas9 cutting generally), so amplified regions look essential regardless of gene function; calling hits first yields false essentials at ERBB2\u002FMYC\u002FFGFR1. Run CRISPRcleanR\u002FChronos BEFORE hit calling, or use CRISPRi to bypass the DSB. This is a pipeline step, not a post-hoc interpretation. Verifying `abs(rho(LFC,CN)) \u003C 0.1` afterwards needs a matched CN profile: CRISPRcleanR corrects unsupervised without one, so the check happens in crispr-screens\u002Fcopy-number-correction (or use Chronos, which consumes CN directly).\n4. **CEGv2 essential-gene depletion is the screen's built-in positive control.** If known essentials do not deplete (CEGv2 PR-AUC below ~0.7), the screen failed selection and NO novel hit is trustworthy regardless of its p-value — the seam analog of a spike-in. Normalize -> QC -> (CN correct) -> hit-call, never hit-call first; add batch as an MLE covariate, never pre-corrected with ComBat on counts (distorts the NB mean-variance the caller assumes).\n\n## Made-once commitments\n\n| Commitment | Consequence inherited downstream |\n|------------|----------------------------------|\n| Guide library (guide->gene map + NTC\u002FCEGv2\u002FNEGv1 control classes) | The counting denominator, the QC calibrator, the hit-calling priors; a missing\u002Fmisassigned class breaks FDR or PR-AUC |\n| Baseline (plasmid pool \u002F Day-0 \u002F vehicle) | Every LFC; drug-vs-Day-0 conflates drug effect with proliferation |\n| Screen type (dropout \u002F enrichment \u002F FACS \u002F drug-modifier) | Which hit-calling method is even valid |\n| Copy-number profile (cancer lines) | Whether amplicon artifacts are removed before hit calling; residual rho(LFC,CN) is the tell |\n\n## Step 0: Confirm Made-Once Commitments\n\n**Before running any code:** if the baseline (plasmid pool \u002F Day-0 \u002F vehicle), library control classes (NTC\u002FCEGv2\u002FNEGv1), screen type (dropout\u002Fenrichment\u002FFACS\u002Fdrug-modifier), or copy-number profile (cancer lines) are not stated, ask the user rather than inferring them from whatever columns happen to be in the file. The governing principle above is not decorative: a wrong-but-silent commitment invalidates every downstream LFC, QC gate, and hit call with no error thrown, and none of Steps 1-7 below can detect it after the fact from the output alone.\n\n## Pipeline Branches by Screen Design\n\n```\n                    Library Design ([[library-design]])\n                              |\n                              v\n                FASTQ Files -> mageck count -> count matrix\n                              |\n                              v\n                Six-Stage QC ([[screen-qc]])\n                              |\n        +---------------------+---------------------+\n        |                                            |\n        v                                            v\n  Cancer cell line?                          Non-cancer?\n  Apply CN correction                        No CN correction needed\n  ([[copy-number-correction]])\n        |                                            |\n        +---------------------+---------------------+\n                              v\n                  Multi-batch? Apply batch covariate\n                  ([[batch-correction]])\n                              |\n                              v\n                 Pick hit-calling method by design ([[hit-calling]])\n                              |\n        +-----------+---------+---------+-----------+-----------+\n        |           |         |         |           |           |\n        v           v         v         v           v           v\n    2-cond       Time      Drug      Essential   Multi-       Specialized\n    MAGeCK RRA   MAGeCK    drugZ     BAGEL2      screen       (PE\u002FBE\u002FSC\u002F\n                 MLE                              JACKS or     in vivo\u002F\n                                                  Chronos      combinat)\n        |           |         |         |           |           |\n        +-----------+---------+---------+-----------+-----------+\n                              v\n                   Tier-based consensus\n                              v\n                Orthogonal validation\n```\n\n## Step 1: Library Design and Pre-Screen Validation\n\nReference [[library-design]] for full library composition. Verify before sequencing:\n- Plasmid pool Gini \u003C0.1 (Li W et al 2015 MAGeCK-VISPR, Genome Biol 16:281)\n- >=99% guides detected at >25 reads\u002Fguide\n- Skew (p90\u002Fp10) \u003C2\n- NTCs comprise ~1% of library; CEGv2 reference essentials + NEGv1 non-essentials included\n\n## Step 2: Guide Counting\n\n**Goal:** Turn raw FASTQ into a per-guide count matrix with consistent sample labels.\n\n**Approach:** Run mageck count with the library CSV, sample labels in column order, the vector adapter trimmed off the 5' end, and median normalization.\n\n```bash\nmageck count \\\n    --list-seq library.csv \\\n    --sample-label Plasmid,Day0,Veh_r1,Veh_r2,Drug_r1,Drug_r2 \\\n    --fastq Plasmid.fq.gz Day0.fq.gz Veh_r1.fq.gz Veh_r2.fq.gz Drug_r1.fq.gz Drug_r2.fq.gz \\\n    --norm-method median \\\n    --output-prefix experiment \\\n    --trim-5 5   # integer base-count (or AUTO), NOT an adapter sequence; 5 trims the CACCG scaffold\n```\n\nFor Cas12a libraries (Inzolia, in4mer): see [[combinatorial-screens]]. For 10X single-cell direct capture: use cellranger-arc or pertpy-aware counting; see [[perturb-seq-analysis]].\n\n## Step 3: Six-Stage Quality Control\n\n**Goal:** Decide whether the screen is analyzable before calling any hits, using six orthogonal QC stages.\n\n**Approach:** Load the count matrix, compute per-sample Gini, zero-fraction, and depth plus replicate correlation against the hard gates below. Essential-gene recovery (CEGv2 PR-AUC) is a separate check computed once endpoint-vs-baseline LFCs exist (it needs CEGv2\u002FNEGv1 labels) -- see screen-qc.\n\n```python\nimport re\nfrom itertools import combinations\nimport pandas as pd\nimport numpy as np\n\ncounts = pd.read_csv('experiment.count.txt', sep='\\t', index_col=0)\ngenes = counts['Gene']\ncount_matrix = counts.drop('Gene', axis=1)\n\ndef gini(x):\n    x = np.sort(x[x > 0].astype(float))\n    if x.size == 0:\n        return np.nan\n    n = x.size\n    cumx = np.cumsum(x)\n    return (n + 1 - 2 * np.sum(cumx) \u002F cumx[-1]) \u002F n\n\nper_sample = pd.DataFrame({\n    'pct_zero': (count_matrix == 0).sum() \u002F len(count_matrix) * 100,\n    'gini': count_matrix.apply(gini),\n    'reads_per_sgrna': count_matrix.sum() \u002F len(count_matrix),\n})\n\nlog_counts = np.log10(count_matrix + 1)\npearson = log_counts.corr()\n\n# Replicate Pearson must average only WITHIN-CONDITION replicate pairs. Averaging every\n# off-diagonal pair -- as a naive `pearson.values[pearson.values \u003C 1].mean()` does --\n# mixes in baseline-vs-endpoint pairs and understates true replicate concordance: on\n# real HAP1 TKOv3 data (T0 + T18_A\u002FB\u002FC) the naive formula gives 0.677 while the true\n# T18-vs-T18 replicate correlation is 0.789, an 0.11 gap straddling the 0.8 QC floor.\n# Group samples by condition (label with any trailing replicate marker -- _r1, _rep2,\n# _A, _2, ... -- stripped off), then average only same-condition pairs.\ndef condition(sample_name):\n    return re.sub(r'_(?:[Rr]ep|[Rr])?\\d+$|_[A-Z]$', '', sample_name)\n\ngroups = {}\nfor col in count_matrix.columns:\n    groups.setdefault(condition(col), []).append(col)\n\nreplicate_pairs = [pearson.loc[a, b] for cols in groups.values() for a, b in combinations(cols, 2)]\n\nprint(per_sample)\nif replicate_pairs:\n    print('Replicate Pearson (within-condition pairs only):', np.mean(replicate_pairs))\nelse:\n    print('Replicate Pearson: no condition has >=2 samples -- cannot assess replicate concordance')\n```\n\nHard gates from [[screen-qc]]:\n- Plasmid Gini \u003C0.1; endpoint \u003C0.3 (or \u003C0.55 for heavy drug screens)\n- Replicate Pearson on log-counts >0.85\n- CEGv2 PR-AUC >0.7 against the Hart 2017 reference essential gene set (community convention, not a threshold defined in that paper)\n- Reads per sgRNA per sample >=300 (DepMap convention)\n\n## Step 4: Copy-Number Correction (Cancer Cell Lines Only)\n\nIf screening in a cancer cell line, apply CRISPRcleanR (unsupervised, no CN profile needed) or Chronos (joint with CN profile). Required to remove Aguirre 2016 \u002F Munoz 2016 amplicon artifact.\n\n**Goal:** Strip the copy-number amplicon artifact that makes amplified regions look essential in cancer lines.\n\n**Approach:** Run CRISPRcleanR unsupervised genome-wide LFC correction (no CN profile needed), then feed the corrected counts downstream; for DepMap-scale panels with matched CN, use Chronos instead.\n\n```r\nlibrary(CRISPRcleanR)\ndata(KY_Library_v1.0)\nnorm \u003C- ccr.NormfoldChanges('experiment.count.txt', min_reads = 30, EXPname = 'screen',\n                              libraryAnnotation = KY_Library_v1.0)   # arg 1 is the file PATH\ngw_lfc \u003C- ccr.logFCs2chromPos(norm$logFCs, KY_Library_v1.0)          # $logFCs, not $norm_fold_changes\ncleaned \u003C- ccr.GWclean(gw_lfc, display = TRUE, label = 'screen')\ncorrected_counts \u003C- ccr.correctCounts('screen', norm$norm_counts, cleaned,\n                                        KY_Library_v1.0)              # (CL, normalised_counts, correctedFCs, libraryAnnotation)\n# ccr.correctCounts returns an in-memory frame; it does NOT write this file. Persist it, because the\n# hit callers below read a count TABLE from disk -- the CN-correction commitment in rule 3 is only\n# honored if that file, not the raw experiment.count.txt, is what MAGeCK \u002F BAGEL2 \u002F drugZ consume.\nwrite.table(corrected_counts, 'screen_cleanr_corrected_counts.txt',\n            sep = '\\t', quote = FALSE, row.names = FALSE)\n```\n\nFor DepMap-scale panels with longitudinal data + matched CN, use Chronos. See [[copy-number-correction]].\n\n## Step 5: Batch Correction (Multi-Batch Screens)\n\nFor multi-batch screens, add batch as a covariate in MAGeCK MLE rather than pre-correcting with ComBat. See [[batch-correction]] for full decision tree.\n\n## Step 6: Method-Matched Hit Calling\n\n**Goal:** Call hits with the method that matches the experimental design, plus at least one orthogonal method for consensus.\n\n**Approach:** Pick by design - RRA or BAGEL2 for two-condition essentiality, MLE for time course, drugZ for drug-modifier, JACKS for multi-screen, Chronos for cancer panels - and run two methods so the consensus step has something to reconcile.\n\n### 6a. Two-condition essentiality (MAGeCK RRA or BAGEL2)\n\nCancer cell lines: pass the CRISPRcleanR-corrected count file (`screen_cleanr_corrected_counts.txt` from Step 4) as `--count-table`\u002F`-i` below, NOT the raw `experiment.count.txt` — the CN-correction commitment is only honored if the corrected counts are what the hit caller reads.\n\nMatch `--treatment-id`\u002F`--control-id` (or BAGEL2 `-c`) to whatever sample labels the count step actually used. The two worked conventions below are both valid — don't mix their column names.\n\nIf your count table uses Step 2's own drug-screen labels (`Plasmid,Day0,Veh_r1,Veh_r2,Drug_r1,Drug_r2`):\n\n```bash\nmageck test \\\n    --count-table experiment.count.txt \\\n    --treatment-id Veh_r1,Veh_r2 \\\n    --control-id Day0 \\\n    --norm-method median \\\n    --output-prefix essentiality_rra\n```\n\nIf your count table instead uses a Day0\u002FDay14 two-timepoint dropout convention (a single endpoint vs Day0 — not the multi-timepoint time-course design in Step 6b, which needs MLE):\n\n```bash\nmageck test \\\n    --count-table experiment.count.txt \\\n    --treatment-id Day14_r1,Day14_r2,Day14_r3 \\\n    --control-id Day0 \\\n    --norm-method median \\\n    --output-prefix essentiality_rra\n```\n\nBAGEL2's `bf` step is non-deterministic unless seeded (`-s`) — build 115's `-s`\u002F`--seed` default is clock-based, and identical reruns on real HAP1 TKOv3 data produced a mean absolute BF difference of 1.31 (max 89.3) across 18,053 genes, flipping 39 genes' `bagel_hit` (BF>6) call between runs. Always pass a fixed `-s`; see [[bagel-essentiality]]'s \"Reproducibility: Fixing the Random Seed\" section for the full verification.\n\n```bash\nBAGEL.py fc -i experiment.count.txt -o experiment -c Day0 --min-reads 30   # -o is an output LABEL; fc writes experiment.foldchange\nBAGEL.py bf -i experiment.foldchange -o bayes_factor.txt -e CEGv2.txt -n NEGv1.txt \\\n    -c Veh_r1,Veh_r2 -s 42   # or -c Day14_r1,Day14_r2,Day14_r3 for the Day0\u002FDay14 convention; add -b -NB 1000 for bootstrapping; -k is not a bf option; -s fixes the seed for reproducible BF\n```\n\n### 6b. Time-course \u002F multi-condition (MAGeCK MLE)\n\n```bash\nmageck mle --count-table experiment.count.txt --design-matrix design.txt \\\n    --output-prefix timecourse_mle --norm-method median\n```\n\n### 6c. Drug-modifier (drugZ)\n\n```bash\npython drugz.py \\\n    -i experiment.count.txt \\\n    -o drugz_output.txt \\\n    -c Veh_r1,Veh_r2 \\\n    -x Drug_r1,Drug_r2 \\\n    -p 5\n```\n\ndrugZ requires vehicle as control, not Day-0. See [[drugz-chemogenomic]].\n\n### 6d. Multi-screen joint analysis (JACKS)\n\n```bash\npython run_JACKS.py experiment.count.txt replicatemap.txt guidemap.txt \\\n    --rep_hdr Replicate --sample_hdr Sample --ctrl_sample_hdr Control \\\n    --sgrna_hdr sgRNA --gene_hdr Gene --outprefix jacks_out --apply_w_hp\n```\n\n### 6e. Cancer cell-line panels (Chronos)\n\n```python\nimport chronos\n# All three inputs must be dicts of DataFrame keyed by library name, not bare DataFrames.\nmodel = chronos.Chronos(sequence_map={'screen': sequence_map},\n                          guide_gene_map={'screen': guide_gene_map},\n                          readcounts={'screen': counts_df})   # readcounts=, not reads=\nmodel.train(nepochs=301)                                 # nepochs (default 301), not n_steps\ngene_effects = model.gene_effect                         # attribute, not a method call\n# Copy-number correction is a separate post-hoc step (chronos.alternate_CN(gene_effect, copy_number) \u002F a CN matrix), not a constructor arg\n```\n\nDepMap quarterly standard; handles CN bias + screen quality + longitudinal jointly.\n\n## Step 7: Tier-Based Consensus\n\n**Goal:** Consolidate the per-method calls into confidence tiers.\n\n**Approach:** Merge each method's gene-level result, threshold each to a per-method hit flag, and tier by how many methods agree (Tier 1 = all three, Tier 2 = two of three). BAGEL2's `bagel_bf` input MUST come from a seeded (`-s`) run (Step 6a) — an unseeded rerun changes which genes cross BF>6 and destabilizes this Tier assignment itself, not just BAGEL2's own output.\n\n```python\nmageck = pd.read_csv('essentiality_rra.gene_summary.txt', sep='\\t')[['id', 'neg|fdr']].rename(\n    columns={'id': 'gene', 'neg|fdr': 'mageck_neg_fdr'})\nbagel = pd.read_csv('bayes_factor.txt', sep='\\t')[['GENE', 'BF']].rename(\n    columns={'GENE': 'gene', 'BF': 'bagel_bf'})\ndrugz_df = pd.read_csv('drugz_output.txt', sep='\\t')[['GENE', 'fdr_synth']].rename(\n    columns={'GENE': 'gene', 'fdr_synth': 'drugz_synth_fdr'})\n\nmerged = mageck.merge(bagel, on='gene', how='outer').merge(drugz_df, on='gene', how='outer')\nmerged['mageck_hit'] = merged['mageck_neg_fdr'] \u003C 0.05\nmerged['bagel_hit'] = merged['bagel_bf'] > 6\nmerged['drugz_hit'] = merged['drugz_synth_fdr'] \u003C 0.05\nmerged['tier'] = merged[['mageck_hit', 'bagel_hit', 'drugz_hit']].astype(int).sum(axis=1)\ntier1 = merged[merged['tier'] >= 3]\ntier2 = merged[merged['tier'] == 2]\nmerged.to_csv('tier_consensus.csv', index=False)   # the documented deliverable; the frame above is otherwise in-memory only\n```\n\n## Specialized Branches\n\n| Screen design | Specialized workflow |\n|---------------|----------------------|\n| Single-cell Perturb-seq \u002F CROP-seq \u002F Multiome | [[perturb-seq-analysis]] -- Pertpy + Mixscape + SCEPTRE |\n| Combinatorial paralog (Cas12a Inzolia \u002F Big Papi) | [[combinatorial-screens]] -- GI scoring; synthetic-lethal identification |\n| Base-editor variant-function (Hanna 2021 style) | [[base-editing-analysis]] + [[crispresso-editing]] |\n| Prime-editor variant installation | [[prime-editing-screens]] -- PRIDICT2 pegRNA design |\n| In vivo tumor \u002F immune screens | [[in-vivo-screens]] -- focused library; per-animal meta-analysis |\n\n## Visualization\n\n**Goal:** Show the hit landscape as a volcano of effect size against significance.\n\n**Approach:** Plot log2 fold change against -log10(FDR), highlight genes past the FDR gate, and save the figure to file.\n\n```python\nimport matplotlib.pyplot as plt\nfig, ax = plt.subplots(figsize=(10, 8))\ngene_summary = pd.read_csv('essentiality_rra.gene_summary.txt', sep='\\t')\nsig = gene_summary['neg|fdr'] \u003C 0.05\nax.scatter(gene_summary.loc[~sig, 'neg|lfc'],\n            -np.log10(gene_summary.loc[~sig, 'neg|fdr'].clip(lower=1e-10)),\n            c='lightgray', alpha=0.5, s=10)\nax.scatter(gene_summary.loc[sig, 'neg|lfc'],\n            -np.log10(gene_summary.loc[sig, 'neg|fdr'].clip(lower=1e-10)),\n            c='red', alpha=0.7, s=18)\nax.axhline(-np.log10(0.05), ls='--', c='black', lw=0.5)\nax.set_xlabel('Log2 Fold Change')\nax.set_ylabel('-Log10(FDR)')\nplt.savefig('volcano.png', dpi=150)\n```\n\nMAGeCKFlute R package provides one-shot FluteRRA \u002F FluteMLE dashboards with KEGG\u002FReactome enrichment.\n\n## Output Files\n\n| File | Source step | Description |\n|------|-------------|-------------|\n| experiment.count.txt | mageck count | Raw count matrix |\n| experiment.countsummary.txt | mageck count | Per-sample Gini, mapping, % zero |\n| screen_cleanr_corrected_counts.txt | CRISPRcleanR | CN-corrected counts (cancer lines) |\n| essentiality_rra.gene_summary.txt | mageck test | Gene-level RRA scores |\n| bayes_factor.txt | BAGEL2 | Per-gene Bayes factors |\n| drugz_output.txt | drugZ | sumZ, normZ, per-direction FDR |\n| jacks_out_gene_JACKS_results.txt | JACKS | Gene effect + sgRNA efficacy |\n| tier_consensus.csv | Custom aggregation | Tier-1\u002F2\u002F3 hits across methods |\n\n## Common Errors\n\n| Symptom | Cause | Fix |\n|---------|-------|-----|\n| False essentials at ERBB2\u002FMYC\u002FFGFR1 | Hit calling before copy-number correction (gene-independent DNA-damage arrest at amplicons, regardless of p53 status) | Run CRISPRcleanR\u002FChronos BEFORE hit calling; verify abs(rho(LFC,CN)) \u003C 0.1; or use CRISPRi to bypass the DSB |\n| FDR broken or PR-AUC uncomputable | NTC (null) and CEGv2 essential (positive control) classes swapped or one absent | Keep both classes; NTCs calibrate the null\u002FFDR, CEGv2\u002FNEGv1 calibrate PR-AUC and BAGEL2\u002FChronos priors |\n| Every hit rescaled \u002F drug effect confounded | Wrong baseline (Day-0 for a drug screen) | Drug screen -> vehicle control; plasmid pool for the cloning-bottleneck baseline |\n| \"Everything significant at FDR\u003C0.01\" | Heavy selection breaks median normalization (>40% guides change) | Switch to `--norm-method control` on NTCs, or BAGEL2 |\n| Underpowered \u002F method mismatch | RRA on a time course; single-line Chronos | Pick method by design (fork table); RRA fails multi-condition, Chronos is overkill single-line |\n| Distorted NB mean-variance | Batch pre-corrected with ComBat on counts | Add batch as a MAGeCK MLE covariate instead |\n| Novel hits from a failed screen | CEGv2 essentials did not deplete (PR-AUC \u003C 0.7) | The screen failed selection; no hit is trustworthy regardless of p-value |\n| Tier consensus changes between identical reruns | BAGEL2 `bf` run unseeded (clock-based default `-s`) | Always pass a fixed `-s` (Step 6a); see [[bagel-essentiality]]'s Reproducibility section |\n\n## References\n\n- Li W, Xu H, Xiao T, et al (2014) MAGeCK enables robust identification of essential genes from genome-scale CRISPR\u002FCas9 knockout screens. *Genome Biology* 15:554. DOI 10.1186\u002Fs13059-014-0554-4.\n- Aguirre AJ, Meyers RM, Weir BA, et al (2016) Genomic copy number dictates a gene-independent cell response to CRISPR\u002FCas9 targeting. *Cancer Discovery* 6:914-929. DOI 10.1158\u002F2159-8290.CD-16-0154. (the amplicon artifact.)\n- Munoz DM, Cassiani PJ, Li L, et al (2016) CRISPR screens provide a comprehensive assessment of cancer vulnerabilities but generate false-positive hits for highly amplified genomic regions. *Cancer Discovery* 6:900-913. DOI 10.1158\u002F2159-8290.CD-16-0178.\n- Hart T, Moffat J (2016) BAGEL: a computational framework for identifying essential genes from pooled library screens. *BMC Bioinformatics* 17:164. DOI 10.1186\u002Fs12859-016-1015-8.\n- Iorio F, Behan FM, Goncalves E, et al (2018) Unsupervised correction of gene-independent cell responses to CRISPR-Cas9 targeting (CRISPRcleanR). *BMC Genomics* 19:604. DOI 10.1186\u002Fs12864-018-4989-y.\n- Joung J, Konermann S, Gootenberg JS, et al (2017) Genome-scale CRISPR-Cas9 knockout and transcriptional activation screening. *Nature Protocols* 12:828-863. DOI 10.1038\u002Fnprot.2017.016. (library QC: skew, zero-count and coverage conventions.)\n\n## Related Skills\n\n- crispr-screens\u002Flibrary-design - Library composition and design rules\n- crispr-screens\u002Fscreen-qc - Six-stage QC + CEGv2 PR-AUC\n- crispr-screens\u002Fmageck-analysis - MAGeCK RRA + MLE detail\n- crispr-screens\u002Fbagel-essentiality - BAGEL2 Bayes factor essentiality\n- crispr-screens\u002Fdrugz-chemogenomic - drugZ for drug-modifier screens\n- crispr-screens\u002Fjacks-analysis - Joint multi-screen analysis with shared efficacy\n- crispr-screens\u002Fhit-calling - Cross-method decision tree + reconciliation\n- crispr-screens\u002Fcopy-number-correction - CRISPRcleanR \u002F CERES \u002F Chronos\n- crispr-screens\u002Fbatch-correction - Multi-batch design matrix\n- crispr-screens\u002Fcrispresso-editing - CRISPResso2 editing quantification\n- crispr-screens\u002Fbase-editing-analysis - Variant-function BE screens\n- crispr-screens\u002Fprime-editing-screens - PRIDICT2 pegRNA design\n- crispr-screens\u002Fperturb-seq-analysis - Single-cell screen analysis\n- crispr-screens\u002Fcombinatorial-screens - Cas12a multiplex + GI scoring\n- crispr-screens\u002Fin-vivo-screens - Bottleneck-aware in vivo design\n- pathway-analysis\u002Fgo-enrichment - Functional enrichment of hits\n- pathway-analysis\u002Fgsea - Pre-ranked GSEA on hit lists\n","## Version Compatibility\n\nReference examples tested with: MAGeCK 0.5.9+, BAGEL2 1.0.5+, drugZ Aug 2019+, JACKS 0.2.0+, Chronos 2.0+, CRISPRcleanR 3.0+ (R), Pertpy 0.6+, PRIDICT2, CRISPResso2 2.2.14+, MAGeCKFlute 2.0+, pandas 2.2+, numpy 1.26+, matplotlib 3.8+.\n\nBefore using code patterns, verify installed versions match. If versions differ:\n- CLI: `mageck --version`, `BAGEL.py fc --help`, `drugz -h`, `CRISPResso --version`\n- Python: `pip show pertpy scanpy anndata` (mageck-vispr via conda `mageck --version`; JACKS\u002FChronos are GitHub installs — check their repos)\n- R: `packageVersion('CRISPRcleanR')`, `packageVersion('MAGeCKFlute')`\n\nIf code throws ImportError, AttributeError, or TypeError, introspect the installed package and adapt the example to match the actual API rather than retrying.\n\n## CRISPR Screen Pipeline\n\n**\"Analyze my pooled or single-cell CRISPR screen end-to-end\"** -> Pick the screen design branch, run guide counting, audit six QC stages, apply copy-number and batch correction as needed, run the design-matched hit-calling method, and consolidate across methods for high-confidence hits.\n\nThis is a workflow skill: it owns the chaining decisions and hand-offs, not the internals of any one step.\n\n## The governing principle\n\nEvery LFC, QC gate, and hit call is computed against a reference that is committed once at library-order time; a wrong-but-silent commitment invalidates the endpoint with no error thrown.\n\n1. **The guide LIBRARY definition (guide->gene map + control classes) is the denominator, the calibrator, and the training reference — committed once.** The library must carry non-targeting controls (NTCs, ~1%, the null distribution) AND CEGv2 reference essentials + NEGv1 non-essentials (the positive\u002Fnegative calibrators for PR-AUC and BAGEL2\u002FChronos priors). NTCs calibrate the null\u002FFDR; CEGv2\u002FNEGv1 calibrate PR-AUC — swapping or dropping a class silently breaks FDR or QC.\n2. **The baseline choice has a right answer and rescales every hit.** Dropout\u002Fenrichment LFC is against a baseline: plasmid pool for the cloning-bottleneck baseline, Day-0\u002FT0 for the biology baseline, and vehicle (NOT Day-0) for drug screens — drug-vs-Day-0 conflates drug effect with normal proliferation.\n3. **Copy-number correction MUST precede hit calling in cancer cell lines.** Multiple simultaneous Cas9 cuts at amplified loci trigger a gene-independent DNA-damage\u002FG2 arrest (Aguirre 2016; Munoz 2016; the effect appears in both TP53-mutant and TP53-wild-type lines, though Aguirre 2016 found TP53 status correlates with its magnitude -- separately, Ihry 2018 \u002F Haapaniemi 2018 report p53-dependent toxicity of Cas9 cutting generally), so amplified regions look essential regardless of gene function; calling hits first yields false essentials at ERBB2\u002FMYC\u002FFGFR1. Run CRISPRcleanR\u002FChronos BEFORE hit calling, or use CRISPRi to bypass the DSB. This is a pipeline step, not a post-hoc interpretation. Verifying `abs(rho(LFC,CN)) \u003C 0.1` afterwards needs a matched CN profile: CRISPRcleanR corrects unsupervised without one, so the check happens in crispr-screens\u002Fcopy-number-correction (or use Chronos, which consumes CN directly).\n4. **CEGv2 essential-gene depletion is the screen's built-in positive control.** If known essentials do not deplete (CEGv2 PR-AUC below ~0.7), the screen failed selection and NO novel hit is trustworthy regardless of its p-value — the seam analog of a spike-in. Normalize -> QC -> (CN correct) -> hit-call, never hit-call first; add batch as an MLE covariate, never pre-corrected with ComBat on counts (distorts the NB mean-variance the caller assumes).\n\n## Made-once commitments\n\n| Commitment | Consequence inherited downstream |\n|------------|----------------------------------|\n| Guide library (guide->gene map + NTC\u002FCEGv2\u002FNEGv1 control classes) | The counting denominator, the QC calibrator, the hit-calling priors; a missing\u002Fmisassigned class breaks FDR or PR-AUC |\n| Baseline (plasmid pool \u002F Day-0 \u002F vehicle) | Every LFC; drug-vs-Day-0 conflates drug effect with proliferation |\n| Screen type (dropout \u002F enrichment \u002F FACS \u002F drug-modifier) | Which hit-calling method is even valid |\n| Copy-number profile (cancer lines) | Whether amplicon artifacts are removed before hit calling; residual rho(LFC,CN) is the tell |\n\n## Step 0: Confirm Made-Once Commitments\n\n**Before running any code:** if the baseline (plasmid pool \u002F Day-0 \u002F vehicle), library control classes (NTC\u002FCEGv2\u002FNEGv1), screen type (dropout\u002Fenrichment\u002FFACS\u002Fdrug-modifier), or copy-number profile (cancer lines) are not stated, ask the user rather than inferring them from whatever columns happen to be in the file. The governing principle above is not decorative: a wrong-but-silent commitment invalidates every downstream LFC, QC gate, and hit call with no error thrown, and none of Steps 1-7 below can detect it after the fact from the output alone.\n\n## Pipeline Branches by Screen Design\n\n```\n                    Library Design ([[library-design]])\n                              |\n                              v\n                FASTQ Files -> mageck count -> count matrix\n                              |\n                              v\n                Six-Stage QC ([[screen-qc]])\n                              |\n        +---------------------+---------------------+\n        |                                            |\n        v                                            v\n  Cancer cell line?                          Non-cancer?\n  Apply CN correction                        No CN correction needed\n  ([[copy-number-correction]])\n        |                                            |\n        +---------------------+---------------------+\n                              v\n                  Multi-batch? Apply batch covariate\n                  ([[batch-correction]])\n                              |\n                              v\n                 Pick hit-calling method by design ([[hit-calling]])\n                              |\n        +-----------+---------+---------+-----------+-----------+\n        |           |         |         |           |           |\n        v           v         v         v           v           v\n    2-cond       Time      Drug      Essential   Multi-       Specialized\n    MAGeCK RRA   MAGeCK    drugZ     BAGEL2      screen       (PE\u002FBE\u002FSC\u002F\n                 MLE                              JACKS or     in vivo\u002F\n                                                  Chronos      combinat)\n        |           |         |         |           |           |\n        +-----------+---------+---------+-----------+-----------+\n                              v\n                   Tier-based consensus\n                              v\n                Orthogonal validation\n```\n\n## Step 1: Library Design and Pre-Screen Validation\n\nReference [[library-design]] for full library composition. Verify before sequencing:\n- Plasmid pool Gini \u003C0.1 (Li W et al 2015 MAGeCK-VISPR, Genome Biol 16:281)\n- >=99% guides detected at >25 reads\u002Fguide\n- Skew (p90\u002Fp10) \u003C2\n- NTCs comprise ~1% of library; CEGv2 reference essentials + NEGv1 non-essentials included\n\n## Step 2: Guide Counting\n\n**Goal:** Turn raw FASTQ into a per-guide count matrix with consistent sample labels.\n\n**Approach:** Run mageck count with the library CSV, sample labels in column order, the vector adapter trimmed off the 5' end, and median normalization.\n\n```bash\nmageck count \\\n    --list-seq library.csv \\\n    --sample-label Plasmid,Day0,Veh_r1,Veh_r2,Drug_r1,Drug_r2 \\\n    --fastq Plasmid.fq.gz Day0.fq.gz Veh_r1.fq.gz Veh_r2.fq.gz Drug_r1.fq.gz Drug_r2.fq.gz \\\n    --norm-method median \\\n    --output-prefix experiment \\\n    --trim-5 5   # integer base-count (or AUTO), NOT an adapter sequence; 5 trims the CACCG scaffold\n```\n\nFor Cas12a libraries (Inzolia, in4mer): see [[combinatorial-screens]]. For 10X single-cell direct capture: use cellranger-arc or pertpy-aware counting; see [[perturb-seq-analysis]].\n\n## Step 3: Six-Stage Quality Control\n\n**Goal:** Decide whether the screen is analyzable before calling any hits, using six orthogonal QC stages.\n\n**Approach:** Load the count matrix, compute per-sample Gini, zero-fraction, and depth plus replicate correlation against the hard gates below. Essential-gene recovery (CEGv2 PR-AUC) is a separate check computed once endpoint-vs-baseline LFCs exist (it needs CEGv2\u002FNEGv1 labels) -- see screen-qc.\n\n```python\nimport re\nfrom itertools import combinations\nimport pandas as pd\nimport numpy as np\n\ncounts = pd.read_csv('experiment.count.txt', sep='\\t', index_col=0)\ngenes = counts['Gene']\ncount_matrix = counts.drop('Gene', axis=1)\n\ndef gini(x):\n    x = np.sort(x[x > 0].astype(float))\n    if x.size == 0:\n        return np.nan\n    n = x.size\n    cumx = np.cumsum(x)\n    return (n + 1 - 2 * np.sum(cumx) \u002F cumx[-1]) \u002F n\n\nper_sample = pd.DataFrame({\n    'pct_zero': (count_matrix == 0).sum() \u002F len(count_matrix) * 100,\n    'gini': count_matrix.apply(gini),\n    'reads_per_sgrna': count_matrix.sum() \u002F len(count_matrix),\n})\n\nlog_counts = np.log10(count_matrix + 1)\npearson = log_counts.corr()\n\n# Replicate Pearson must average only WITHIN-CONDITION replicate pairs. Averaging every\n# off-diagonal pair -- as a naive `pearson.values[pearson.values \u003C 1].mean()` does --\n# mixes in baseline-vs-endpoint pairs and understates true replicate concordance: on\n# real HAP1 TKOv3 data (T0 + T18_A\u002FB\u002FC) the naive formula gives 0.677 while the true\n# T18-vs-T18 replicate correlation is 0.789, an 0.11 gap straddling the 0.8 QC floor.\n# Group samples by condition (label with any trailing replicate marker -- _r1, _rep2,\n# _A, _2, ... -- stripped off), then average only same-condition pairs.\ndef condition(sample_name):\n    return re.sub(r'_(?:[Rr]ep|[Rr])?\\d+$|_[A-Z]$', '', sample_name)\n\ngroups = {}\nfor col in count_matrix.columns:\n    groups.setdefault(condition(col), []).append(col)\n\nreplicate_pairs = [pearson.loc[a, b] for cols in groups.values() for a, b in combinations(cols, 2)]\n\nprint(per_sample)\nif replicate_pairs:\n    print('Replicate Pearson (within-condition pairs only):', np.mean(replicate_pairs))\nelse:\n    print('Replicate Pearson: no condition has >=2 samples -- cannot assess replicate concordance')\n```\n\nHard gates from [[screen-qc]]:\n- Plasmid Gini \u003C0.1; endpoint \u003C0.3 (or \u003C0.55 for heavy drug screens)\n- Replicate Pearson on log-counts >0.85\n- CEGv2 PR-AUC >0.7 against the Hart 2017 reference essential gene set (community convention, not a threshold defined in that paper)\n- Reads per sgRNA per sample >=300 (DepMap convention)\n\n## Step 4: Copy-Number Correction (Cancer Cell Lines Only)\n\nIf screening in a cancer cell line, apply CRISPRcleanR (unsupervised, no CN profile needed) or Chronos (joint with CN profile). Required to remove Aguirre 2016 \u002F Munoz 2016 amplicon artifact.\n\n**Goal:** Strip the copy-number amplicon artifact that makes amplified regions look essential in cancer lines.\n\n**Approach:** Run CRISPRcleanR unsupervised genome-wide LFC correction (no CN profile needed), then feed the corrected counts downstream; for DepMap-scale panels with matched CN, use Chronos instead.\n\n```r\nlibrary(CRISPRcleanR)\ndata(KY_Library_v1.0)\nnorm \u003C- ccr.NormfoldChanges('experiment.count.txt', min_reads = 30, EXPname = 'screen',\n                              libraryAnnotation = KY_Library_v1.0)   # arg 1 is the file PATH\ngw_lfc \u003C- ccr.logFCs2chromPos(norm$logFCs, KY_Library_v1.0)          # $logFCs, not $norm_fold_changes\ncleaned \u003C- ccr.GWclean(gw_lfc, display = TRUE, label = 'screen')\ncorrected_counts \u003C- ccr.correctCounts('screen', norm$norm_counts, cleaned,\n                                        KY_Library_v1.0)              # (CL, normalised_counts, correctedFCs, libraryAnnotation)\n# ccr.correctCounts returns an in-memory frame; it does NOT write this file. Persist it, because the\n# hit callers below read a count TABLE from disk -- the CN-correction commitment in rule 3 is only\n# honored if that file, not the raw experiment.count.txt, is what MAGeCK \u002F BAGEL2 \u002F drugZ consume.\nwrite.table(corrected_counts, 'screen_cleanr_corrected_counts.txt',\n            sep = '\\t', quote = FALSE, row.names = FALSE)\n```\n\nFor DepMap-scale panels with longitudinal data + matched CN, use Chronos. See [[copy-number-correction]].\n\n## Step 5: Batch Correction (Multi-Batch Screens)\n\nFor multi-batch screens, add batch as a covariate in MAGeCK MLE rather than pre-correcting with ComBat. See [[batch-correction]] for full decision tree.\n\n## Step 6: Method-Matched Hit Calling\n\n**Goal:** Call hits with the method that matches the experimental design, plus at least one orthogonal method for consensus.\n\n**Approach:** Pick by design - RRA or BAGEL2 for two-condition essentiality, MLE for time course, drugZ for drug-modifier, JACKS for multi-screen, Chronos for cancer panels - and run two methods so the consensus step has something to reconcile.\n\n### 6a. Two-condition essentiality (MAGeCK RRA or BAGEL2)\n\nCancer cell lines: pass the CRISPRcleanR-corrected count file (`screen_cleanr_corrected_counts.txt` from Step 4) as `--count-table`\u002F`-i` below, NOT the raw `experiment.count.txt` — the CN-correction commitment is only honored if the corrected counts are what the hit caller reads.\n\nMatch `--treatment-id`\u002F`--control-id` (or BAGEL2 `-c`) to whatever sample labels the count step actually used. The two worked conventions below are both valid — don't mix their column names.\n\nIf your count table uses Step 2's own drug-screen labels (`Plasmid,Day0,Veh_r1,Veh_r2,Drug_r1,Drug_r2`):\n\n```bash\nmageck test \\\n    --count-table experiment.count.txt \\\n    --treatment-id Veh_r1,Veh_r2 \\\n    --control-id Day0 \\\n    --norm-method median \\\n    --output-prefix essentiality_rra\n```\n\nIf your count table instead uses a Day0\u002FDay14 two-timepoint dropout convention (a single endpoint vs Day0 — not the multi-timepoint time-course design in Step 6b, which needs MLE):\n\n```bash\nmageck test \\\n    --count-table experiment.count.txt \\\n    --treatment-id Day14_r1,Day14_r2,Day14_r3 \\\n    --control-id Day0 \\\n    --norm-method median \\\n    --output-prefix essentiality_rra\n```\n\nBAGEL2's `bf` step is non-deterministic unless seeded (`-s`) — build 115's `-s`\u002F`--seed` default is clock-based, and identical reruns on real HAP1 TKOv3 data produced a mean absolute BF difference of 1.31 (max 89.3) across 18,053 genes, flipping 39 genes' `bagel_hit` (BF>6) call between runs. Always pass a fixed `-s`; see [[bagel-essentiality]]'s \"Reproducibility: Fixing the Random Seed\" section for the full verification.\n\n```bash\nBAGEL.py fc -i experiment.count.txt -o experiment -c Day0 --min-reads 30   # -o is an output LABEL; fc writes experiment.foldchange\nBAGEL.py bf -i experiment.foldchange -o bayes_factor.txt -e CEGv2.txt -n NEGv1.txt \\\n    -c Veh_r1,Veh_r2 -s 42   # or -c Day14_r1,Day14_r2,Day14_r3 for the Day0\u002FDay14 convention; add -b -NB 1000 for bootstrapping; -k is not a bf option; -s fixes the seed for reproducible BF\n```\n\n### 6b. Time-course \u002F multi-condition (MAGeCK MLE)\n\n```bash\nmageck mle --count-table experiment.count.txt --design-matrix design.txt \\\n    --output-prefix timecourse_mle --norm-method median\n```\n\n### 6c. Drug-modifier (drugZ)\n\n```bash\npython drugz.py \\\n    -i experiment.count.txt \\\n    -o drugz_output.txt \\\n    -c Veh_r1,Veh_r2 \\\n    -x Drug_r1,Drug_r2 \\\n    -p 5\n```\n\ndrugZ requires vehicle as control, not Day-0. See [[drugz-chemogenomic]].\n\n### 6d. Multi-screen joint analysis (JACKS)\n\n```bash\npython run_JACKS.py experiment.count.txt replicatemap.txt guidemap.txt \\\n    --rep_hdr Replicate --sample_hdr Sample --ctrl_sample_hdr Control \\\n    --sgrna_hdr sgRNA --gene_hdr Gene --outprefix jacks_out --apply_w_hp\n```\n\n### 6e. Cancer cell-line panels (Chronos)\n\n```python\nimport chronos\n# All three inputs must be dicts of DataFrame keyed by library name, not bare DataFrames.\nmodel = chronos.Chronos(sequence_map={'screen': sequence_map},\n                          guide_gene_map={'screen': guide_gene_map},\n                          readcounts={'screen': counts_df})   # readcounts=, not reads=\nmodel.train(nepochs=301)                                 # nepochs (default 301), not n_steps\ngene_effects = model.gene_effect                         # attribute, not a method call\n# Copy-number correction is a separate post-hoc step (chronos.alternate_CN(gene_effect, copy_number) \u002F a CN matrix), not a constructor arg\n```\n\nDepMap quarterly standard; handles CN bias + screen quality + longitudinal jointly.\n\n## Step 7: Tier-Based Consensus\n\n**Goal:** Consolidate the per-method calls into confidence tiers.\n\n**Approach:** Merge each method's gene-level result, threshold each to a per-method hit flag, and tier by how many methods agree (Tier 1 = all three, Tier 2 = two of three). BAGEL2's `bagel_bf` input MUST come from a seeded (`-s`) run (Step 6a) — an unseeded rerun changes which genes cross BF>6 and destabilizes this Tier assignment itself, not just BAGEL2's own output.\n\n```python\nmageck = pd.read_csv('essentiality_rra.gene_summary.txt', sep='\\t')[['id', 'neg|fdr']].rename(\n    columns={'id': 'gene', 'neg|fdr': 'mageck_neg_fdr'})\nbagel = pd.read_csv('bayes_factor.txt', sep='\\t')[['GENE', 'BF']].rename(\n    columns={'GENE': 'gene', 'BF': 'bagel_bf'})\ndrugz_df = pd.read_csv('drugz_output.txt', sep='\\t')[['GENE', 'fdr_synth']].rename(\n    columns={'GENE': 'gene', 'fdr_synth': 'drugz_synth_fdr'})\n\nmerged = mageck.merge(bagel, on='gene', how='outer').merge(drugz_df, on='gene', how='outer')\nmerged['mageck_hit'] = merged['mageck_neg_fdr'] \u003C 0.05\nmerged['bagel_hit'] = merged['bagel_bf'] > 6\nmerged['drugz_hit'] = merged['drugz_synth_fdr'] \u003C 0.05\nmerged['tier'] = merged[['mageck_hit', 'bagel_hit', 'drugz_hit']].astype(int).sum(axis=1)\ntier1 = merged[merged['tier'] >= 3]\ntier2 = merged[merged['tier'] == 2]\nmerged.to_csv('tier_consensus.csv', index=False)   # the documented deliverable; the frame above is otherwise in-memory only\n```\n\n## Specialized Branches\n\n| Screen design | Specialized workflow |\n|---------------|----------------------|\n| Single-cell Perturb-seq \u002F CROP-seq \u002F Multiome | [[perturb-seq-analysis]] -- Pertpy + Mixscape + SCEPTRE |\n| Combinatorial paralog (Cas12a Inzolia \u002F Big Papi) | [[combinatorial-screens]] -- GI scoring; synthetic-lethal identification |\n| Base-editor variant-function (Hanna 2021 style) | [[base-editing-analysis]] + [[crispresso-editing]] |\n| Prime-editor variant installation | [[prime-editing-screens]] -- PRIDICT2 pegRNA design |\n| In vivo tumor \u002F immune screens | [[in-vivo-screens]] -- focused library; per-animal meta-analysis |\n\n## Visualization\n\n**Goal:** Show the hit landscape as a volcano of effect size against significance.\n\n**Approach:** Plot log2 fold change against -log10(FDR), highlight genes past the FDR gate, and save the figure to file.\n\n```python\nimport matplotlib.pyplot as plt\nfig, ax = plt.subplots(figsize=(10, 8))\ngene_summary = pd.read_csv('essentiality_rra.gene_summary.txt', sep='\\t')\nsig = gene_summary['neg|fdr'] \u003C 0.05\nax.scatter(gene_summary.loc[~sig, 'neg|lfc'],\n            -np.log10(gene_summary.loc[~sig, 'neg|fdr'].clip(lower=1e-10)),\n            c='lightgray', alpha=0.5, s=10)\nax.scatter(gene_summary.loc[sig, 'neg|lfc'],\n            -np.log10(gene_summary.loc[sig, 'neg|fdr'].clip(lower=1e-10)),\n            c='red', alpha=0.7, s=18)\nax.axhline(-np.log10(0.05), ls='--', c='black', lw=0.5)\nax.set_xlabel('Log2 Fold Change')\nax.set_ylabel('-Log10(FDR)')\nplt.savefig('volcano.png', dpi=150)\n```\n\nMAGeCKFlute R package provides one-shot FluteRRA \u002F FluteMLE dashboards with KEGG\u002FReactome enrichment.\n\n## Output Files\n\n| File | Source step | Description |\n|------|-------------|-------------|\n| experiment.count.txt | mageck count | Raw count matrix |\n| experiment.countsummary.txt | mageck count | Per-sample Gini, mapping, % zero |\n| screen_cleanr_corrected_counts.txt | CRISPRcleanR | CN-corrected counts (cancer lines) |\n| essentiality_rra.gene_summary.txt | mageck test | Gene-level RRA scores |\n| bayes_factor.txt | BAGEL2 | Per-gene Bayes factors |\n| drugz_output.txt | drugZ | sumZ, normZ, per-direction FDR |\n| jacks_out_gene_JACKS_results.txt | JACKS | Gene effect + sgRNA efficacy |\n| tier_consensus.csv | Custom aggregation | Tier-1\u002F2\u002F3 hits across methods |\n\n## Common Errors\n\n| Symptom | Cause | Fix |\n|---------|-------|-----|\n| False essentials at ERBB2\u002FMYC\u002FFGFR1 | Hit calling before copy-number correction (gene-independent DNA-damage arrest at amplicons, regardless of p53 status) | Run CRISPRcleanR\u002FChronos BEFORE hit calling; verify abs(rho(LFC,CN)) \u003C 0.1; or use CRISPRi to bypass the DSB |\n| FDR broken or PR-AUC uncomputable | NTC (null) and CEGv2 essential (positive control) classes swapped or one absent | Keep both classes; NTCs calibrate the null\u002FFDR, CEGv2\u002FNEGv1 calibrate PR-AUC and BAGEL2\u002FChronos priors |\n| Every hit rescaled \u002F drug effect confounded | Wrong baseline (Day-0 for a drug screen) | Drug screen -> vehicle control; plasmid pool for the cloning-bottleneck baseline |\n| \"Everything significant at FDR\u003C0.01\" | Heavy selection breaks median normalization (>40% guides change) | Switch to `--norm-method control` on NTCs, or BAGEL2 |\n| Underpowered \u002F method mismatch | RRA on a time course; single-line Chronos | Pick method by design (fork table); RRA fails multi-condition, Chronos is overkill single-line |\n| Distorted NB mean-variance | Batch pre-corrected with ComBat on counts | Add batch as a MAGeCK MLE covariate instead |\n| Novel hits from a failed screen | CEGv2 essentials did not deplete (PR-AUC \u003C 0.7) | The screen failed selection; no hit is trustworthy regardless of p-value |\n| Tier consensus changes between identical reruns | BAGEL2 `bf` run unseeded (clock-based default `-s`) | Always pass a fixed `-s` (Step 6a); see [[bagel-essentiality]]'s Reproducibility section |\n\n## References\n\n- Li W, Xu H, Xiao T, et al (2014) MAGeCK enables robust identification of essential genes from genome-scale CRISPR\u002FCas9 knockout screens. *Genome Biology* 15:554. DOI 10.1186\u002Fs13059-014-0554-4.\n- Aguirre AJ, Meyers RM, Weir BA, et al (2016) Genomic copy number dictates a gene-independent cell response to CRISPR\u002FCas9 targeting. *Cancer Discovery* 6:914-929. DOI 10.1158\u002F2159-8290.CD-16-0154. (the amplicon artifact.)\n- Munoz DM, Cassiani PJ, Li L, et al (2016) CRISPR screens provide a comprehensive assessment of cancer vulnerabilities but generate false-positive hits for highly amplified genomic regions. *Cancer Discovery* 6:900-913. DOI 10.1158\u002F2159-8290.CD-16-0178.\n- Hart T, Moffat J (2016) BAGEL: a computational framework for identifying essential genes from pooled library screens. *BMC Bioinformatics* 17:164. DOI 10.1186\u002Fs12859-016-1015-8.\n- Iorio F, Behan FM, Goncalves E, et al (2018) Unsupervised correction of gene-independent cell responses to CRISPR-Cas9 targeting (CRISPRcleanR). *BMC Genomics* 19:604. DOI 10.1186\u002Fs12864-018-4989-y.\n- Joung J, Konermann S, Gootenberg JS, et al (2017) Genome-scale CRISPR-Cas9 knockout and transcriptional activation screening. *Nature Protocols* 12:828-863. DOI 10.1038\u002Fnprot.2017.016. (library QC: skew, zero-count and coverage conventions.)\n\n## Related Skills\n\n- crispr-screens\u002Flibrary-design - Library composition and design rules\n- crispr-screens\u002Fscreen-qc - Six-stage QC + CEGv2 PR-AUC\n- crispr-screens\u002Fmageck-analysis - MAGeCK RRA + MLE detail\n- crispr-screens\u002Fbagel-essentiality - BAGEL2 Bayes factor essentiality\n- crispr-screens\u002Fdrugz-chemogenomic - drugZ for drug-modifier screens\n- crispr-screens\u002Fjacks-analysis - Joint multi-screen analysis with shared efficacy\n- crispr-screens\u002Fhit-calling - Cross-method decision tree + reconciliation\n- crispr-screens\u002Fcopy-number-correction - CRISPRcleanR \u002F CERES \u002F Chronos\n- crispr-screens\u002Fbatch-correction - Multi-batch design matrix\n- crispr-screens\u002Fcrispresso-editing - CRISPResso2 editing quantification\n- crispr-screens\u002Fbase-editing-analysis - Variant-function BE screens\n- crispr-screens\u002Fprime-editing-screens - PRIDICT2 pegRNA design\n- crispr-screens\u002Fperturb-seq-analysis - Single-cell screen analysis\n- crispr-screens\u002Fcombinatorial-screens - Cas12a multiplex + GI scoring\n- crispr-screens\u002Fin-vivo-screens - Bottleneck-aware in vivo design\n- pathway-analysis\u002Fgo-enrichment - Functional enrichment of hits\n- pathway-analysis\u002Fgsea - Pre-ranked GSEA on hit lists\n",{"name":5,"workflow":24,"tool_type":25,"depends_on":26,"description":7,"primary_tool":42,"qc_checkpoints":43},true,"mixed",[27,28,29,30,31,32,33,34,35,36,37,38,39,40,41],"crispr-screens\u002Flibrary-design","crispr-screens\u002Fscreen-qc","crispr-screens\u002Fmageck-analysis","crispr-screens\u002Fbagel-essentiality","crispr-screens\u002Fdrugz-chemogenomic","crispr-screens\u002Fjacks-analysis","crispr-screens\u002Fhit-calling","crispr-screens\u002Fcopy-number-correction","crispr-screens\u002Fbatch-correction","crispr-screens\u002Fcrispresso-editing","crispr-screens\u002Fbase-editing-analysis","crispr-screens\u002Fprime-editing-screens","crispr-screens\u002Fperturb-seq-analysis","crispr-screens\u002Fcombinatorial-screens","crispr-screens\u002Fin-vivo-screens","MAGeCK",[44,46,48,50],{"after_counting":45},">65% mapping rate; \u003C0.5% zero-count in plasmid; Gini \u003C0.1 on plasmid",{"after_qc":47},"Replicate Pearson on log-counts >=0.8 (MAGeCK-VISPR floor; >0.85 acceptable, >0.95 ideal); Spearman >0.7; CEGv2 PR-AUC >0.7",{"after_cn_correction":49},"Spearman ρ between CN and gene LFC abs \u003C0.10 post-correction (literature 'significant bias' band; \u003C0.05 is a stricter target). Requires a matched CN profile, which the unsupervised CRISPRcleanR path never loads -- compute in crispr-screens\u002Fcopy-number-correction, or use Chronos, which takes CN as input",{"after_hit_calling":51},"Tier-1 hits = 3-method consensus; Tier-2 = 2 of 3; Tier-3 = single-method exploratory","main","e0956dc9606356ab78468936b502e55a365282e2","specialists\u002Fcrispr-screen-analyst\u002Fversions\u002F1.0.0\u002Fpackage\u002Fskills\u002Fbio-workflows-crispr-screen-pipeline",0,"https:\u002F\u002Fgithub.com\u002Fmrsonord2240\u002Fopenscience-specialists","https:\u002F\u002Fgithub.com\u002Fmrsonord2240","2026-10-12 03:19:04",{"source":8,"skillName":5,"commands":60,"prompt":70,"zipAvailable":24},[61,64,67],{"tool":62,"command":63},"npx","npx skills add mrsonord2240\u002Fopenscience-specialists --skill bio-workflows-crispr-screen-pipeline",{"tool":65,"command":66},"bunx","bunx skills add mrsonord2240\u002Fopenscience-specialists --skill bio-workflows-crispr-screen-pipeline",{"tool":68,"command":69},"pnpm","pnpm dlx skills add mrsonord2240\u002Fopenscience-specialists --skill bio-workflows-crispr-screen-pipeline","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fmrsonord2240\u002Fopenscience-specialists\u002Fblob\u002Fmain\u002Fspecialists\u002Fcrispr-screen-analyst\u002Fversions\u002F1.0.0\u002Fpackage\u002Fskills\u002Fbio-workflows-crispr-screen-pipeline\u002FSKILL.md",[72,78],{"path":73,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":76,"sourceHtmlUrl":77},"examples\u002Fcrispr_pipeline.sh","SCRIPT","脚本",1722,"https:\u002F\u002Fgithub.com\u002Fmrsonord2240\u002Fopenscience-specialists\u002Fblob\u002Fmain\u002Fspecialists\u002Fcrispr-screen-analyst\u002Fversions\u002F1.0.0\u002Fpackage\u002Fskills\u002Fbio-workflows-crispr-screen-pipeline\u002Fexamples\u002Fcrispr_pipeline.sh",{"path":79,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":82,"sourceHtmlUrl":83},"usage-guide.md","REFERENCE","参考文档",11918,"https:\u002F\u002Fgithub.com\u002Fmrsonord2240\u002Fopenscience-specialists\u002Fblob\u002Fmain\u002Fspecialists\u002Fcrispr-screen-analyst\u002Fversions\u002F1.0.0\u002Fpackage\u002Fskills\u002Fbio-workflows-crispr-screen-pipeline\u002Fusage-guide.md",[85,92,99,106,113,120],{"skillNo":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"publishedAt":19},"01M7CW36A6QAGWFE2C008M1XPG","bio-workflows-metabolomics-pipeline","Orchestrates the untargeted LC-MS metabolomics pipeline end-to-end (xcms 4.x feature extraction, QC\u002Fdrift\u002Fnormalization, confidence-stratified annotation, permutation-validated statistics, background-aware pathway mapping), naming what each stage decides and where it silently fails. Use when running a full LC-MS metabolomics study from raw mzML to enriched pathways and needing the honest handoffs between stages. Each stage defers to its component skill for parameters and traps; for stable-isotope flux (a separate branch, not this untargeted flow) see metabolomics\u002Fisotope-tracing.","https:\u002F\u002Fgithub.com\u002Fmrsonord2240\u002Fopenscience-specialists\u002Fblob\u002Fmain\u002Fspecialists\u002Funtargeted-metabolomics-analyst\u002Fversions\u002F1.0.0\u002Fpackage\u002Fskills\u002Fbio-workflows-metabolomics-pipeline\u002FSKILL.md",[],[],{"skillNo":93,"name":94,"nameZh":6,"description":95,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":98,"publishedAt":19},"01M7CW2Y18623NA2S1Y30FRQW1","bio-metabolomics-xcms-preprocessing","Programmatic untargeted LC-MS feature extraction in R with the modern xcms 4.x MsExperiment\u002FXcmsExperiment API, taking raw mzML to a feature table via CentWave peak detection, retention-time alignment, peak-density correspondence, gap-filling, CAMERA redundancy collapse, and built-in QC feature filtering. Use when converting centroided LC-MS runs into a features-by-samples matrix and deciding centWave\u002Fgrouping\u002Falignment parameters. For drift correction and QC\u002FCV filtering execution see metabolomics\u002Fnormalization-qc; for metabolite identification see metabolomics\u002Fmetabolite-annotation; for the MS-DIAL GUI alternative with MS2Dec deconvolution see metabolomics\u002Fmsdial-preprocessing; for downstream statistics see metabolomics\u002Fstatistical-analysis.","https:\u002F\u002Fgithub.com\u002Fmrsonord2240\u002Fopenscience-specialists\u002Fblob\u002Fmain\u002Fspecialists\u002Funtargeted-metabolomics-analyst\u002Fversions\u002F1.0.0\u002Fpackage\u002Fskills\u002Fbio-metabolomics-xcms-preprocessing\u002FSKILL.md",[],[],{"skillNo":100,"name":101,"nameZh":6,"description":102,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":103,"categories":104,"keywords":105,"publishedAt":19},"01M7CW2M364HVWPNQEKE2WP4PP","bio-metabolomics-statistical-analysis","Decision-grade statistical analysis for metabolomics intensity tables. Covers transformation and scaling (Pareto vs unit-variance as a hidden hypothesis), unsupervised structure (PCA\u002FHCA for QC), permutation-validated PLS-DA\u002FOPLS-DA (R2 vs Q2, double CV, VIP as heuristic), univariate testing (Welch\u002FMann-Whitney\u002FANOVA\u002FLMM with covariate adjustment), and dependence-aware multiple testing. Use when testing which metabolites differ, building or validating a discriminant model, choosing a scaling, or correcting many correlated tests. For sample-wise normalization\u002Fdrift correction see metabolomics\u002Fnormalization-qc; for ML classifiers and selection-inside-CV leakage see machine-learning\u002Fbiomarker-discovery and machine-learning\u002Fmodel-validation; for pathway interpretation see metabolomics\u002Fpathway-mapping; for design\u002Fpower\u002Fmultiplicity regime see experimental-design\u002Fmultiple-testing.","https:\u002F\u002Fgithub.com\u002Fmrsonord2240\u002Fopenscience-specialists\u002Fblob\u002Fmain\u002Fspecialists\u002Funtargeted-metabolomics-analyst\u002Fversions\u002F1.0.0\u002Fpackage\u002Fskills\u002Fbio-metabolomics-statistical-analysis\u002FSKILL.md",[],[],{"skillNo":107,"name":108,"nameZh":6,"description":109,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":110,"categories":111,"keywords":112,"publishedAt":19},"01M7CW2BCBF851WYE03Z8XPEDQ","bio-metabolomics-pathway-mapping","Maps metabolomics results to biological pathways via over-representation (ORA), metabolite-set enrichment (MSEA\u002FQEA), mummichog\u002FPSEA on raw m\u002Fz peaks, and network-diffusion enrichment (FELLA), with correct background-set construction and honest interpretive ceilings. Use when interpreting differential metabolites or an untargeted LC-MS feature table in pathway context, choosing ORA vs MSEA vs mummichog vs topology, or setting the reference\u002Fbackground set. For annotation confidence levels feeding ORA see metabolomics\u002Fmetabolite-annotation; for gene-set concepts see pathway-analysis\u002Fgo-enrichment and pathway-analysis\u002Fgsea; for joint gene+metabolite pathways see multi-omics-integration\u002Fmofa-integration.","https:\u002F\u002Fgithub.com\u002Fmrsonord2240\u002Fopenscience-specialists\u002Fblob\u002Fmain\u002Fspecialists\u002Funtargeted-metabolomics-analyst\u002Fversions\u002F1.0.0\u002Fpackage\u002Fskills\u002Fbio-metabolomics-pathway-mapping\u002FSKILL.md",[],[],{"skillNo":114,"name":115,"nameZh":6,"description":116,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":117,"categories":118,"keywords":119,"publishedAt":19},"01M7CW21NSZMCPMF7D9CHJM718","bio-metabolomics-normalization-qc","Designs QC, corrects signal drift, removes batch effects, filters features, normalizes samples, and imputes missing values for untargeted LC-MS\u002FGC-MS metabolomics, framing each step as a measurement model that can create or erase biological signal. Use when processing a peak\u002Ffeature table before statistical analysis, choosing a drift-correction or sample-normalization method, deciding QC RSD vs D-ratio filtering, or handling left-censored missing values. The feature table is produced by metabolomics\u002Fxcms-preprocessing or metabolomics\u002Fmsdial-preprocessing; transformation\u002Fscaling for modeling defers to metabolomics\u002Fstatistical-analysis; cross-study design issues link to experimental-design\u002Fbatch-design.","https:\u002F\u002Fgithub.com\u002Fmrsonord2240\u002Fopenscience-specialists\u002Fblob\u002Fmain\u002Fspecialists\u002Funtargeted-metabolomics-analyst\u002Fversions\u002F1.0.0\u002Fpackage\u002Fskills\u002Fbio-metabolomics-normalization-qc\u002FSKILL.md",[],[],{"skillNo":121,"name":122,"nameZh":6,"description":123,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":124,"categories":125,"keywords":126,"publishedAt":19},"01M7CW1SWVM0NYHVE6Q7MA3G7R","bio-metabolomics-msdial-preprocessing","Runs the MS-DIAL preprocessing workflow (peak picking, MS2Dec spectral deconvolution, alignment, gap-filling) and imports the alignment-result table into R or Python with honest filtering. Use when preprocessing LC-MS DDA\u002FDIA (SWATH) or GC-MS raw data with MS-DIAL, deciding MS-DIAL vs XCMS, configuring the MSDIALCUI console run, or parsing an MS-DIAL export into a clean feature matrix. For programmatic R peak detection and the feature-table-as-artifact framing see metabolomics\u002Fxcms-preprocessing; for lipid annotation mode see metabolomics\u002Flipidomics; for MSI-level confidence honesty see metabolomics\u002Fmetabolite-annotation; for drift correction and QC see metabolomics\u002Fnormalization-qc.","https:\u002F\u002Fgithub.com\u002Fmrsonord2240\u002Fopenscience-specialists\u002Fblob\u002Fmain\u002Fspecialists\u002Funtargeted-metabolomics-analyst\u002Fversions\u002F1.0.0\u002Fpackage\u002Fskills\u002Fbio-metabolomics-msdial-preprocessing\u002FSKILL.md",[],[]]