Claude Skills by Pavel-Kravchenko

Probability for Bioinformatics with NumPy

R quick-reference for bioinformatics — syntax differences from Python, vectors, data frames, Bioconductor patterns, and statistical distributions

R hypothesis testing and nonparametric methods: binomial test, sign test, Wilcoxon, Kruskal-Wallis, and power analysis. Use when choosing and applying statistical tests in R.

R regression, correlation, and diagnostics: confidence intervals, t-tests, normality testing, Pearson/Spearman/Kendall correlation, Fisher z-transform CI.

Statistics with Python

Generative imaging and inverse-problem restoration workflows.

DNA/RNA sequence foundation models: embeddings, fine-tuning, and regulatory prediction.

DNA tokenization strategies, k-mer baselines, and pretrained sequence embeddings for genomic ML.

Route prioritized genomic variants into AlphaFold2/3 or RoseTTAFold structural follow-up.

Graph algorithms (BFS, DFS, Dijkstra, MST, topological sort) and dynamic programming (memoization, tabulation, knapsack, sequence alignment)

V(D)J repertoire analysis, HLA typing, and neoantigen prediction pipelines.

Linked lists, stacks, queues, dynamic arrays, BST/AVL/Red-Black trees, hash tables with collision resolution, and Bloom filters

Linux shell commands, git workflows, bash scripting, and file encoding handling for bioinformatics data processing.

Oxford Nanopore and PacBio long-read sequencing — basecalling (Dorado), QC (NanoStat), alignment (Minimap2), assembly (Flye, Hifiasm), SV calling (Sniffles2), methylation, isoform analysis

Metabolomics analysis workflow: LC-MS data processing, metabolite identification, statistical analysis, and pathway mapping. Use when conducting metabolomics experiments.

16S/amplicon metagenomics diversity and community analysis.

Shotgun metagenomics — host decontamination, Kraken2 taxonomic profiling, Bracken abundance, HUMAnN3 functional annotation, MEGAHIT assembly, MetaBAT2 binning, CheckM, MAGs, QIIME2 16S

Biological network analysis — PPI networks from STRING, NetworkX graph metrics, Louvain community detection, GO enrichment per module, Cytoscape export, gene regulatory network inference with GENIE3

Population structure correction and stratification-aware QC for association studies.

ESM2 embeddings, ESMFold structure prediction, zero-shot mutation scoring, and protein design.

OOP for bioinformatics classes, decorators, context managers, error handling, and SQL queries for biological databases.

Python classes for bioinformatics — designing Sequence, Gene, and ProteinRecord classes with encapsulation, dunder methods, inheritance, ABCs, and properties.

Python comprehensions and generators for bioinformatics — list/dict/set comprehensions, generator expressions for large sequence files, bio-specific patterns.

Context managers for safe file, database, and resource handling in bioinformatics pipelines.

Python control flow patterns specific to bioinformatics — codon iteration, frame scanning, motif search, sequence classification.

Python data types for bioinformatics: int, float, str, bool, None, and type conversions with biological examples. Use when learning how Python represents biological data.

Matplotlib and seaborn for bioinformatics figures: volcano plots, heatmaps, MA plots, genome tracks. Critical gotchas for publication-quality figures.

Pandas patterns for bio data — missing values, duplicates, type conversion, wide/long reshaping, string parsing, apply/transform/pipe.

Python decorator patterns for bioinformatics: timing, validation, memoization, and stacking.

Dictionaries and sets for bioinformatics: codon tables, nucleotide frequency, defaultdict grouping, Counter k-mers, and set operations on gene lists.

Robust error handling for bioinformatics pipelines — custom exception hierarchy, try/except/else/finally, exception chaining, strict FASTA parsing.

Python arithmetic, comparison, logical, and membership operators with bioinformatics applications — codon frames, GC content, quality filters.

FASTA/FASTQ/CSV parsing recipes, streaming file I/O patterns, and format-specific pitfalls for bioinformatics data files

Bio-specific function patterns — ORF finding, sequence analysis, mutable default trap, and functional programming idioms for bioinformatics

Generator patterns for streaming bioinformatics data — FASTA/FASTQ readers, sliding windows, translation pipelines, and memory-efficient sequence processing

Custom iterators for bioinformatics — codon iteration, k-mer generation, streaming FASTA/FASTQ, and lazy sequence processing patterns

List and tuple patterns for bioinformatics — codon splitting, gene coordinate records, named tuples, sorting by GC/length, and copy pitfalls

NumPy for bioinformatics — vectorized operations, broadcasting for expression matrices, PWM construction, sliding-window GC with cumsum, and normalization recipes

OOP patterns for bioinformatics — subscriptable sequence databases, sliceable BioSeq, callable motif scorers, __slots__ for millions of variants, and composable mixins

Operator pitfalls and idioms for bioinformatics — floor division for codons, modulo for reading frames, GC content precedence, and membership testing

NumPy arrays and Pandas DataFrames for bioinformatics — vectorized ops, broadcasting, CPM/RPKM normalization, PWM scoring, loc/iloc, groupby, and annotation merges.

Python Introduction for Bioinformatics

Regular expressions for biological sequence pattern matching — ORF finding, motifs, restriction sites, FASTA headers, overlapping matches

Python string operations for biological sequences: reverse complement, codon extraction, motif scanning, FASTA parsing. Bio-specific patterns and pitfalls.

Sets and Counter for bioinformatics: gene list operations (intersection/union/difference), k-mer deduplication, and unique element tracking. Bio-specific patterns only.

Ensembl, UCSC Genome Browser, NCBI, and dbSNP are all backed by relational databases. Even locally, SQLite is a practical way to store and query gene annotations, variant tables, and expression result

Python string operations for bioinformatics: reverse complement, GC content, motif finding, FASTA parsing. Bio-specific patterns and pitfalls only.

Lists and tuples for bioinformatics: codon extraction, genomic coordinate records, named tuples, and copy semantics. Bio-specific patterns only.

Python variables and data types for bioinformatics — naming, mutability, references, numeric and string types

Python collections (namedtuple, defaultdict, Counter, set) and regex for bioinformatics — k-mer counting, FASTA/FASTQ streaming, restriction maps, PROSITE patterns.