테스트·보안 스킬

6 production-ready AI engineering workflows: prompt evaluation (8-dimension scoring), context budget planning, RAG pipeline design, agent security audit (65-point checklist), eval harness building, and product sense coaching.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

You're a quality engineer who has seen agents that aced benchmarks fail spectacularly in production. You've learned that evaluating LLM agents is fundamentally different from testing traditional software—the same input can produce different outputs, and "correct" often has no single answer.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Compares Hi-C contact maps between conditions across the right scale -- differential bin-pair contacts (multiHiCcompare, diffHic), differential A/B compartments (dcHiC), differential TAD boundaries (delta insulation), and differential loops (diffloop, DiffHiChIP) -- with distance-stratified between-sample normalization, replicate-aware NB-GLM FDR, HiCRep SCC reproducibility gating, and CNV correction for cancer/aneuploid samples. Use when comparing Hi-C between treatment and control, finding differential contacts/compartments/boundaries/loops, normalizing two maps of unequal depth, choosing a replicate-aware test, gating replicates with SCC, or correcting copy-number artifacts before a tumor-vs-normal comparison.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Performs set operations on genomic intervals - intersect (-wa/-wb/-wo/-wao/-loj/-c/-v/-u), subtract (-A), merge (-d, -c/-o), complement, cluster, multiinter, unionbedg, map, and groupby - with bedtools (CLI) and pybedtools/pyranges/bioframe (Python). Covers the sorted-input contract and the -sorted chromosome-order footgun, reciprocal/fractional overlap (-f/-F/-r/-e) and the A-vs-B asymmetry, -split for spliced/BED12/BAM features, and jaccard/fisher as mechanics only. Use when finding overlapping or unique regions between BED/peak/feature files, building consensus peaksets, removing blacklisted regions, transferring annotation values onto intervals, or computing interval-set similarity; route overlap-significance testing to overlap-significance.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Analyze spatial cell-cell interactions, neighborhoods, and niches in IMC/MIBI data with squidpy and imcRtools, covering neighborhood-enrichment permutation nulls, the abundance-vs-density confound, inhomogeneous Ripley's K, cellular-neighborhood discovery, graph-construction (contact vs proximity), and edge effects. Use when testing whether cell types co-locate, choosing a spatial null, building a neighbor graph, discovering tissue niches, or deciding whether a spatial pattern is real or a density/segmentation artifact.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Compare cell-type composition and spatial features across conditions in IMC/MIBI cohorts with the patient as the experimental unit, covering pseudoreplication, per-patient aggregation, mixed models, compositional (Dirichlet/scCODA) differential abundance, diffcyt, per-image-to-patient spatial differential testing (SpaceANOVA), batch covariates, and FDR. Use when testing whether a cell type or spatial niche differs between groups, avoiding cell-level pseudoreplication, choosing a differential-abundance method, or correctly powering an IMC cohort comparison.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Tests whether two genomic interval sets overlap (colocalize) more than expected by chance using a permutation test against a structured-genome null model. Covers bedtools fisher (analytic 2x2 screen), bedtools shuffle + jaccard permutation, GAT (isochore/GC-conditioned simulation with FDR), regioneR (flexible permutation, randomizeRegions vs circularRandomizeRegions, localZScore), LOLA (universe-relative Fisher against a region database), and GREAT/rGREAT (regulatory-domain binomial + hypergeometric for ontology-from-regions). Stresses the universe/background choice, matched background, blacklist exclusion, and multiple-testing control. Use when asking whether peaks/regions are enriched at enhancers/TFBS/features, scoring region-set colocalization or region-set enrichment, comparing CNV/SV concordance, or turning an overlap count into a defensible p-value.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Handles batch effects in bulk RNA-seq via design-matrix inclusion (the correct path for DE), ComBat/ComBat-seq for visualization, SVA for unknown latent factors, RUVSeq for negative-control-gene-anchored unwanted variation, and limma::removeBatchEffect for plotting only. Encodes the Nygaard 2016 cardinal sin against testing on a batch-corrected matrix, the choice between SVA/RUVg/RUVs/RUVr, the confounding non-identifiability problem, the single-cell boundary (Harmony/MNN are NOT for bulk), and the Goh 2017 harmonization critique. Use when designing a DE analysis with batch structure, troubleshooting batch-dominated PCA, choosing ComBat vs ComBat-seq, handling unknown batch via SVA, integrating across studies, or deciding when (rarely) to subtract batch.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Compare gene co-expression and regulatory networks between biological conditions to find rewired relationships using DiffCorr, DiffCoEx, DINGO/iDINGO, and CoDiNA. Covers the differential-connectivity-is-not-differential-expression distinction, the pairwise multiple-testing explosion, marginal vs partial (direct) rewiring, and the underpowered-rewiring failure mode. Use when comparing co-expression networks between disease vs control, treatment, or developmental stages, or finding hub genes that rewire without changing mean expression. For single-condition modules see coexpression-networks; for differential expression of means see differential-expression/de-results.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Identifies differential m6A methylation between conditions from MeRIP-seq paired IP/input data using exomePeak2 with `bam_ip` + `bam_input` (control arm) and `bam_treated_ip` + `bam_treated_input` (treatment arm) for integrated GC-bias-aware differential calling (Liu 2022 *NAR Genom Bioinform* 4:lqac046), QNB beta-binomial test (Liu 2017 *BMC Bioinformatics* 18:387), MeTDiff HMM-based differential bundled with MeTPeak, RADAR (Zhang 2019 *Genome Biol* 20:294) with its `filterBins -> diffIP -> reportResult` workflow, and the defensible paired-symmetric case of edgeR / DESeq2 on featureCounts-on-peaks matrices when batch / lot covariates need explicit fixed-effect handling (exomePeak2's top-level API does NOT accept arbitrary covariates). Covers paired vs unpaired vs interaction designs, batch confounding (antibody lot, RNA prep, sequencing run) and the per-lot meta-analysis strategy when exomePeak2 is the primary caller, the stoichiometry-vs-expression-vs-IP-efficiency confound that all MeRIP differential methods inherit, normalisation choice (size factor on IP, on input, on per-sample IP/input ratio), the McIntyre 2020 reproducibility caveat, effect-size filtering as a guardrail against under-powered N=2 designs, and orthogonal-validation routes for absolute stoichiometry (GLORI / SAC-seq / m6Anet `mod_ratio`). Use when comparing m6A levels across two or more conditions, choosing between exomePeak2 / QNB / RADAR / MeTDiff for a given design, handling batch confounding when exomePeak2's API is too rigid, normalising against input properly, distinguishing real hyper- / hypo-methylation from expression-level shifts, applying effect-size thresholds, interpreting volcano plots of differential peaks, or planning a follow-up orthogonal stoichiometry validation.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Controls error rates across thousands of simultaneous tests in genomics discovery using false-discovery-rate methods (Benjamini-Hochberg 1995; Benjamini-Yekutieli 2001 for arbitrary dependence; Storey q-value with pi0 estimation; local FDR; independent filtering Bourgon 2010; covariate-weighted FDR via IHW Ignatiadis 2016), plus family-wise error control (Bonferroni, Holm) and the GWAS genome-wide threshold. Covers the FDR-versus-FWER choice as the discovery-versus-confirmatory distinction, the dependence assumptions behind BH (PRDS) versus BY, pi0 estimation, the independent-filtering and false-coverage-rate traps, and reproducibility ranking via IDR (Li 2011). Use when correcting p-values from genome-wide tests, choosing between BH/BY/q-value/Bonferroni, setting an FDR threshold, applying IHW or independent filtering, or interpreting q-values. For confirmatory trials with few pre-specified endpoints (closed testing, graphical/gatekeeping), see clinical-biostatistics/multiplicity-graphical.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Unsupervised clustering and cell-type identification for high-dimensional flow, spectral, and mass cytometry - FlowSOM, PhenoGraph, FlowSOM-via-CATALYST, with UMAP/tSNE for visualization. Covers the type-vs-state marker distinction (cluster on lineage, test state within clusters), over-provision-then-metacluster, the Weber-Robinson benchmark, seed dependence and metacluster stability, why embeddings are for looking not measuring, and median-heatmap annotation/merging. Use when discovering populations without predefined gates, choosing a clustering algorithm, selecting the number of metaclusters, or annotating clusters into cell types.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Analyzes species-environment relationships with constrained ordination (CCA, RDA, db-RDA), variance partitioning, indicator species (indicspecies IndVal.g group-equalized), PERMANOVA paired MANDATORILY with PERMDISP (Anderson & Walsh 2013; dispersion confounds centroid tests), Joint Species Distribution Models (HMSC, sjSDM, gjam) with explicit rejection of "residual covariance equals biotic interaction", phylogenetic community ecology (SES_MPD/MNTD), trait-environment via RLQ + fourth-corner with corrected modeltype=6 (Dray 2014), bipartite network metrics (NODF, modularity) with curveball null (Strona 2014), and Mantel-test replacements (dbRDA, GDM) for spatial data. Use when testing how environmental gradients structure communities, identifying habitat indicator taxa, partitioning variance among predictors, deciding whether PERMANOVA significance is location vs dispersion, picking among HMSC/sjSDM/gjam, or replacing Mantel tests for landscape data.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Performs differential expression on bulk RNA-seq count data with edgeR's negative-binomial GLM and quasi-likelihood F-test framework. Covers DGEList construction, filterByExpr, TMM/TMMwsp normalization, robust dispersion estimation, glmQLFit/glmQLFTest, TREAT for magnitude-bounded hypotheses, contrasts via no-intercept designs, voom and voomWithQualityWeights for heterogeneous samples, and the edgeR v4 bias-corrected APL changes. Use when running bulk DE with edgeR, choosing edgeR over DESeq2 (small n, transcript DE via catchSalmon, large samples), needing TREAT for a fold-change-threshold hypothesis, troubleshooting v3-to-v4 reproducibility, building paired or interaction designs, or handling library-quality heterogeneity.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Performs differential expression on bulk RNA-seq count data with DESeq2's negative-binomial GLM, Wald and LRT testing, apeglm/ashr/normal LFC shrinkage, independent filtering, Cook's outlier handling, VST/rlog transforms, and design formulas including paired, batch, and interaction terms. Use when running bulk DE, choosing DESeq2 over edgeR or limma-voom, building a paired or interaction design, applying LFC shrinkage for ranking or GSEA, choosing Wald vs LRT, troubleshooting padj=NA, picking VST vs rlog, importing salmon/kallisto via tximport, or analyzing prokaryotic RNA-seq.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Differential abundance (DA) and differential state (DS) analysis for flow and mass cytometry - tests which cell populations change in frequency or marker expression between conditions using diffcyt (edgeR/voom/GLMM for DA, limma/LMM for DS), with cydar, CITRUS, and compositional methods (sccomp, scCODA, DCATS) as alternatives. Covers the sample-is-the-experimental-unit principle, design/contrast and mixed-model formulas, compositionality of cluster proportions, and FDR across clusters. Use when comparing populations between groups, choosing a DA method, handling paired/batch designs, or deciding whether compositional correction is needed.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Delimits putative species boundaries from molecular data within the de Queiroz 2007 unified-lineage framework using ASAP (Puillandre 2021 successor to ABGD), mPTP C++ (Kapli 2017 successor to bPTP; bPTP is Python NOT R), GMYC single/multi-threshold (Pons 2006; Fujisawa 2013), multilocus BPP v4 with prior calibration from data (NOT defaults; Yang 2015), SNAPP + BFD* for SNP delimitation, DELINEATE (Sukumaran 2021) speciation-process modeling to address Sukumaran & Knowles 2017 PNAS critique that MSC delimits structure not species, integrative-taxonomy congruence (Padial 2010; Carstens 2013), Dsuite for introgression testing before sister claims (Malinsky 2021), and Meyer & Paulay 2005 barcoding-gap-absence caveat. Use when delineating species from DNA barcoding data, resolving cryptic complexes, choosing among ASAP/mPTP/BPP/DELINEATE, calibrating BPP priors, distinguishing introgression from ILS, or applying the Sukumaran-Knowles oversplitting correction.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Extracts, filters, annotates, and exports differential expression results from DESeq2 or edgeR with proper handling of padj=NA (independent filtering, Cook's outliers, all-zero), multiple-testing correction choice (BH vs Storey q-value vs IHW vs lfsr), TREAT vs post-hoc fold-change filtering, p-value histogram diagnostics, gene annotation via org.db/biomaRt/mygene, GSEA preranked input, ORA background construction, replication reality (Schurch 2016 small-n result), and SABV/sex-stratified reporting. Use when extracting and interpreting DE results, troubleshooting padj=NA, choosing FDR method, preparing ranked lists for pathway analysis, annotating gene IDs, or comparing DESeq2 vs edgeR outputs.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Estimates time-scaled phylogenies, molecular clock rates, effective reproduction number R_e (or R_t), and population dynamics from dated pathogen genomes using TreeTime (maximum-likelihood) and BEAST2 (Bayesian; strict / uncorrelated lognormal / ORC clocks; constant / exponential / Bayesian Skyline / Skygrid / BICEPS / Birth-Death-Skyline / sampled-ancestor BDSKY priors; structured coalescent via MASCOT). Covers root-to-tip clock signal QC via TempEst, date-randomisation tests (Ramsden 2009; Duchêne 2015), recombination masking via Gubbins and ClonalFrameML before clock inference for recombining bacteria, BDSKY origin-vs-rootHeight pitfalls, sampling-bias correction (Volz & Frost 2014; preferential-sampling extensions), MASCOT structured coalescent for migration, BICEPS-vs-BSP skyline choice, multi-chain BEAST2 convergence diagnostics, and reconciliation between phylodynamic R_e and case-based R_t. Use when dating outbreak origins, estimating substitution rates, inferring R_e through time, building time-calibrated Nextstrain Augur trees, choosing between strict and relaxed clocks, fitting Birth-Death-Skyline (Stadler 2013) models, diagnosing temporal-signal failure, masking recombination before clock inference for *Streptococcus pneumoniae* / *Neisseria gonorrhoeae* / *Klebsiella* / *E. coli* phylodynamics, running MASCOT for structured-population analyses, or using UShER for pandemic-scale placement.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Analyzes pooled CRISPR screens with MAGeCK (Li et al 2014), covering count generation (mageck count), the RRA two-condition workflow (mageck test using alpha-RRA over per-sgRNA negative-binomial p-values), the MLE multi-condition workflow (mageck mle with explicit design matrix and beta-score output), normalization choice (median vs total vs control-sgRNA vs spike-in), sgRNA efficiency injection, paired-sample testing, time-course design, drug-screen versus dropout-screen design matrices, MAGeCKFlute and MAGeCK-VISPR downstream visualization, and decision logic for when to use MAGeCK vs JACKS / BAGEL2 / drugZ / Chronos. Use when running a fresh CRISPR screen analysis, picking RRA vs MLE for the experimental design, choosing a normalization method from QC signatures, debugging MLE convergence failure or NaN beta scores, comparing MAGeCK output across tools, or building a batch-aware multi-cell-line / multi-condition MLE design matrix.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Cross-method decision tree for calling hits in pooled CRISPR screens. Catalogs statistical models (MAGeCK RRA, MAGeCK MLE, BAGEL2, drugZ, JACKS, Chronos, CERES), experimental designs each is built for, failure modes outside design domain, reconciliation when methods disagree, multiple-testing and effect-size thresholds, the order of operations (count -> QC -> CN-correct -> hit-call -> validate), the second-best-sgRNA conservative rule, and consensus-hit strategy. Use when choosing among MAGeCK / BAGEL2 / drugZ / JACKS / Chronos for a given design, reconciling disagreement across two or three methods on the same screen, deciding whether to require consensus, gating downstream validation by hit-confidence tier, or interpreting unstable hit lists across reruns.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Detect positive (diversifying / episodic / pervasive) selection using codon dN/dS frameworks. Implements PAML codeml site models (M0/M1a/M2a/M7/M8/M8a), branch models, branch-site model A (Zhang 2005), and HyPhy methods (BUSTED, BUSTED-S, BUSTED-MH, BUSTED-PH, MEME, FEL, FUBAR, aBSREL, SLAC, RELAX, GARD, FUBAR-MH). Includes McDonald-Kreitman framework (asymptotic alpha, impMKT, polyDFE, DFE-alpha, GRAPES) for within-species + divergence inference, RERconverge for trait-correlated rate shifts, CSUBST for convergent substitution, and PhyloAcc for accelerated noncoding evolution. Use when testing adaptive evolution at codons, branches, or full gene; running GARD recombination pre-screen; controlling alignment-error and gBGC false positives; reconciling PAML vs HyPhy results; or performing genome-scale selection scans.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Reconstruct ancestral states at internal phylogenetic nodes for sequences (PAML codeml, IQ-TREE --ancestral, GRASP, FastML), discrete traits (corHMM hidden-rate Markov, ape::ace, phytools::make.simmap stochastic mapping, BayesTraits), and continuous traits (phytools::fastAnc, geiger Brownian/OU, RPANDA). Use when designing constructs for ancestral protein resurrection, tracing trait evolution along a tree, performing stochastic character mapping, testing models of trait evolution (BM vs OU vs EB), inferring ancestral genome content via Dollo or DTL reconciliation, or quantifying ancestral-state uncertainty for downstream comparative analyses.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Model gene-family birth-death dynamics across a species tree using CAFE5 (Mendes et al 2020 Bioinformatics 36:5516 gamma-distributed rate categories), CAFE5-error (annotation-error-aware), Count (Csurös 2010 ancestral state reconstruction), BadiRate (Librado 2012 likelihood + parsimony), DupliPHY-Family, and ALE/AleRax (for per-family DTL; see [[gene-tree-species-tree-reconciliation]]). Test lineage-specific gene-family expansions and contractions, distinguish biological dynamics from annotation artifacts, account for assembly fragmentation, identify functional enrichment in expanded / contracted families. Use when correlating gene-family changes with phenotype evolution, ranking lineages by adaptive gene-family-rate shifts, post-WGD dosage-balance analysis, or building Birth-death models from OrthoFinder presence/absence matrices.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.