Computational Biology and Data Science

Retrieve paginated biological records from a public API while caching responses and recording accession identifiers.

Design a normalized SQL schema for samples, experimental conditions, measurements, and biological annotations.

Write a SQL query that joins biological tables and computes grouped summaries.

Enforce schema, uniqueness, range, and referential-integrity checks on an analysis dataset.

Profile the runtime and memory use of a data-processing task and select an appropriate chunking strategy.

Express a multi-step bioinformatics analysis as a dependency-aware Snakemake workflow.

Package a computational analysis environment in a container from a version-pinned specification.

Submit a resource-specified batch job to a SLURM cluster and diagnose its completion status.

Summarize the center, spread, skew, and outliers of a biological measurement using robust statistics and plots.

Simulate binomial, Poisson, and normal data and connect each distribution to an appropriate biological process.

Estimate marginal and conditional probabilities from simulated diagnostic or phenotype data.

Update the probability of a biological hypothesis with Bayes’ rule using code and a frequency visualization.

Generate a sampling distribution to explain how sample size affects estimator variability.

Construct and interpret a bootstrap confidence interval for a biological effect.

Fit a simple linear regression model and interpret its slope in biological terms.

Use residual plots to identify nonlinearity, unequal variance, and influential observations in a fitted regression.

Define the observational unit, outcome, comparison, and estimand for a biological research question.

Perform a permutation test for a difference between two biological conditions.

Report an effect estimate and confidence interval separately from a p-value.

Control the false discovery rate in a multiple-testing analysis and interpret adjusted p-values.

Estimate statistical power for a proposed biological comparison by simulation.

Identify confounding, pseudoreplication, data leakage, and batch effects in an experimental plan.

Choose among randomized, blocked, paired, and factorial designs for a stated biological experiment.

Fit a logistic regression model and interpret a predictor’s odds ratio for a binary phenotype.

Validate the structural consistency of FASTA, FASTQ, GFF, and VCF files and explain how their identifiers connect.

Assess raw sequencing-read quality from base-quality, adapter-content, duplication, and length summaries.

Trim adapters and low-quality sequence from reads and verify the effect of trimming with before-and-after quality reports.

Compare local and global pairwise alignments and select the appropriate approach for a stated sequence question.

Run a sequence-similarity search and interpret alignment coverage, identity, E-value, and database context.

Map sequencing reads to a reference genome and inspect coverage and alignment quality at a selected locus.

Call and filter small variants using depth, quality, allele-balance, and missingness criteria.

Generate and evaluate a draft microbial genome assembly using contiguity, completeness, contamination, and read-support evidence.

Annotate a microbial genome and summarize its coding sequences, functional categories, and annotation limitations.

Infer orthologous gene clusters across microbial isolates and construct a gene presence–absence matrix.

Build and interpret a core-genome phylogeny with appropriate rooting and branch-support information.

Measure population structure and genetic relatedness among microbial isolates.

Detect acquired resistance genes and known resistance-associated variants while recording database provenance.

Perform a bacterial genome-wide association test that accounts for microbial population structure.

Construct a machine-learning feature matrix from genomic variants, gene content, phenotype labels, and isolate metadata.

Distinguish causal resistance mechanisms from lineage-linked predictive markers using genomic and experimental evidence.

Create leakage-resistant training and test partitions using biological groups, lineages, or experimental batches.

Build a preprocessing pipeline that learns imputation, encoding, and scaling only from training data.

Train a regularized logistic-regression baseline for a biological classification problem.

Train a tree-based ensemble and compare its error pattern with the linear baseline.

Select and interpret metrics for imbalanced biological classes, including precision, recall, PR-AUC, and ROC-AUC.

Tune model hyperparameters and estimate generalization performance with nested cross-validation.

Assess probability calibration and choose a decision threshold for a stated biological use case.

Use permutation importance or SHAP values to generate cautiously framed biological hypotheses from a fitted model.

Reconcile sample metadata with an assay matrix and flag swapped, duplicated, or unmatched samples.

Normalize bulk RNA-seq count data with a Bioconductor workflow and explain what the normalization corrects.

Diagnose RNA-seq sample quality using library-size summaries, normalized expression distributions, and principal-component plots.

Specify a differential-expression design that includes the biological comparison and relevant batch variables.

Interpret effect sizes, uncertainty, and false-discovery rates in MA and volcano plots.

Perform pathway or gene-set enrichment analysis using an explicitly defined gene universe.

Preprocess an NMR chemical-shift feature matrix while preserving missingness, temperature, and motif metadata.

Evaluate whether a chemotype classifier generalizes across 60°C and 80°C experimental conditions.

Map UniProt, RefSeq, PDB, and predicted-structure identifiers for a microbial protein of interest.

Build a multiple-sequence alignment and identify conserved residues in a microbial protein family.

Extract chains, residues, ligands, secondary structure, and missing regions from a PDB or mmCIF structure.

Assess experimental-structure quality and predicted-structure confidence using method-appropriate metrics.

Create a molecular visualization that communicates an active site, interface, or domain architecture.

Map sequence variants onto a protein structure and classify their proximity to functional features.

Identify candidate pockets or protein–protein interfaces and document the evidence supporting them.

Distinguish structural hypotheses supported by templates or confidence metrics from claims that docking or predicted structures cannot justify.

Translate a target-product goal into measurable chassis, pathway, yield, titer, rate, and process constraints.

Reconstruct a target biosynthetic pathway from curated reaction, enzyme, and metabolite records.

Run flux-balance analysis for a microbial metabolic model and interpret the predicted flux distribution.

Simulate gene deletions or reaction constraints and rank engineering interventions by predicted production phenotype.

Design an expression cassette in silico and verify its orientation, reading frame, regulatory parts, and junctions.

Evaluate codon usage and sequence liabilities for heterologous expression in a selected microbial chassis.

Design a factorial fermentation experiment that separates medium, temperature, aeration, and strain effects.

Fit a response-surface model to fermentation data and propose operating conditions with stated uncertainty.

Annotate a phage genome and distinguish confidently assigned proteins from hypothetical proteins.

Assess phage-genome completeness, terminal-structure evidence, and possible host contamination.

Classify a candidate phage as lytic or temperate using genomic evidence and calibrated uncertainty.

Screen a phage candidate for lysogeny, toxin, virulence, and antimicrobial-resistance liabilities.

Predict candidate bacterial hosts using CRISPR spacers, sequence similarity, and composition-based evidence.

Analyze efficiency-of-plating or killing-curve data to quantify host range and lytic performance.

Compare related phage genomes to identify conserved modules, mosaic regions, and taxonomic placement.

Rank phage candidates for a stated development use case using genomic safety, host-range, lytic-performance, and uncertainty evidence.

Scope an existing AMR or GXM project as a reproducible prediction study with a falsifiable primary question.

Convert a phage-development or precision-fermentation analysis into a scoped companion portfolio case study with a clear decision and deliverable.

Write a data card for the flagship study that documents provenance, cohort composition, measurement conditions, missingness, and reuse constraints.

Write an analysis plan that predefines data splits, baselines, evaluation metrics, and major sensitivity analyses.

Configure an automated, version-pinned workflow that regenerates the flagship study’s principal results.

Create a publication-style figure panel that connects exploratory evidence, model performance, and biological interpretation.

Write a portfolio repository README that documents setup and execution while linking the flagship and companion artifacts.

Deliver a five-minute PhD-style technical talk explaining the research question, methods, evidence, limitations, and next experiment.