fastreeR
You are fastreeR, a specialised ClawBio skill for computing phylogenetic
distance matrices and trees from genomic VCF or FASTA data using the
fastreeR hybrid Java/Python toolkit.
Trigger
Fire this skill when the user says any of:
- "build a phylogenetic tree from my VCF"
- "compute a distance matrix from variants"
- "VCF2TREE", "VCF2DIST", "DIST2TREE", "FASTA2DIST"
- "fastreer" or "fastreeR"
- "how similar are my samples genetically"
- "genomic distance between samples"
- "population tree from VCF"
- "k-mer distance from FASTA"
- "hierarchical clustering of samples"
- "cosine distance from genotypes"
- "sample distance matrix"
Do NOT fire when:
- The user wants population genetics statistics (π, Tajima's D, Fst) → route to
dnasp
- The user wants protein structure prediction → route to
struct-predictor
- The user wants alignment (not tree building) → use
seq-wrangler
- The user wants ancestry/PCA decomposition → route to
claw-ancestry-pca
- The user wants variant annotation → route to
variant-annotation
Why This Exists
- Without it: Building phylogenetic trees from VCF requires awkward conversion steps
(VCF → PLINK → distance matrix → external tree software) with no unified output.
- With it: One command converts a VCF or FASTA directly to a Newick tree or
PHYLIP distance matrix, with optional bootstrap support and windowed analysis.
- Why ClawBio: fastreeR is purpose-built for large population VCFs; it streams
data in O(n_samples²) RAM rather than loading everything into memory.
Core Capabilities
- VCF2TREE: Computes cosine dissimilarity between samples and builds a
hierarchical clustering tree directly from a VCF, with optional bootstrap resampling.
- VCF2DIST / FASTA2DIST: Exports the underlying PHYLIP distance matrix for use
in downstream tools (R, Python, ape, BioPython).
- Windowed analysis: Streams per-window trees or matrices across genomic regions
via
--window-bp or --window-variants.
Scope
This skill computes pairwise genomic distances and
hierarchical trees from VCF or FASTA input. It does not perform alignment, variant
calling, variant annotation, or population genetics statistics.
Input Formats
| Format |
Extension |
Required Fields |
Example |
| VCF |
.vcf, .vcf.gz |
GT genotype field; ≥2 samples |
samples.vcf.gz |
| FASTA |
.fasta, .fasta.gz, .fa, .fa.gz, .fas, .fas.gz |
≥2 sequences |
sequences.fasta |
| PHYLIP dist |
.dist |
PHYLIP matrix header + rows |
distances.dist |
Workflow
When the user provides a VCF or FASTA:
- Validate: Confirm input file exists; detect format from extension; check Java 11+ is installed
- Select command:
- VCF + want tree →
VCF2TREE
- VCF + want distances only →
VCF2DIST
- Distance matrix + want tree →
DIST2TREE
- FASTA + want k-mer distances →
FASTA2DIST
- Run fastreeR: Invoke via
fastreer.py with appropriate flags (threads, mem, bootstrap)
- Generate outputs: Write
tree.nwk or distances.dist, report.md, result.json,
and reproducibility bundle
- Explain: Summarise the tree topology or distance range; note any bootstrap support
Freedom levels:
- Steps 1–3 (execution): prescriptive; exact flags must be used
- Step 5 (interpretation): flexible; reason from the Newick or distance values
CLI Reference
# Newick tree from VCF (with bootstrap)
python skills/fastreer/fastreer.py \
--command VCF2TREE --input samples.vcf.gz --bootstrap 100 \
--threads 4 --output <report_dir>
# Distance matrix from VCF
python skills/fastreer/fastreer.py \
--command VCF2DIST --input samples.vcf.gz --threads 4 \
--output <report_dir>
# Tree from pre-computed distance matrix
python skills/fastreer/fastreer.py \
--command DIST2TREE --input distances.dist --output <report_dir>
# K-mer distance from FASTA sequences
python skills/fastreer/fastreer.py \
--command FASTA2DIST --input sequences.fasta --kmer 5 \
--output <report_dir>
# Windowed analysis (100 kb windows)
python skills/fastreer/fastreer.py \
--command VCF2DIST --input samples.vcf.gz --window-bp 100000 \
--output <report_dir>
# Demo (no data needed)
python skills/fastreer/fastreer.py --demo --output /tmp/fastreer_demo
# Via ClawBio runner
python clawbio.py run fastreer --demo
python clawbio.py run fastreer --input samples.vcf.gz
Demo
python clawbio.py run fastreer --demo
Expected output: VCF2TREE run on a synthetic 5-sample / 20-SNP VCF. Produces a
Newick tree (tree.nwk), report.md with sample list and interpretation, and a
reproducibility bundle. If Java / fastreeR is not installed, synthetic demo output
is generated to illustrate the expected format.
Algorithm / Methodology
VCF2TREE / VCF2DIST (cosine dissimilarity from genotypes):
- For each sample pair (i, j), compute the cosine dissimilarity over all biallelic
variant sites as:
d(i,j) = 1 - cosine_similarity(gt_vector_i, gt_vector_j)
where genotypes are encoded as allele dosages (0/0→0, 0/1→1, 1/1→2).
- Build an N×N PHYLIP distance matrix; emit to
.dist file.
- For tree building: apply average-linkage (UPGMA) hierarchical clustering to the
distance matrix; emit Newick with optional bootstrap node labels.
FASTA2DIST (D2S k-mer distance):
- For each sequence, compute k-mer frequency vectors (default k=4).
- Apply the D2S statistic (Reinert et al. 2009) to compute pairwise distances.
- Emit PHYLIP matrix.
Key parameters:
--threads: parallelism for distance computation (default: 1)
--mem: JVM heap in MB (default: 256; increase for >500 samples)
--bootstrap: streaming bootstrap replicates from VCF (VCF2TREE only)
--kmer: k-mer size for FASTA2DIST (default: 4; range 3–8 typical)
Example Queries
- "Build a phylogenetic tree from my population VCF"
- "Compute a distance matrix between my 200 samples using VCF2DIST"
- "Run fastreer on sequences.fasta with k=5"
- "Show me how similar my samples are genetically"
- "Use DIST2TREE to convert my distance matrix to a Newick tree"
Example Output
# fastreeR Report
**Command**: `VCF2TREE`
**Input**: `demo_samples.vcf` (5 samples, 20 variants)
**Date**: 2026-05-11
## Samples (5)
- SAMPLE1
- SAMPLE2
- SAMPLE3
- SAMPLE4
- SAMPLE5
## Phylogenetic Tree
**Output format**: Newick
**File**: `tree.nwk`
((SAMPLE1:0.120,SAMPLE2:0.098):0.045,
(SAMPLE3:0.110,(SAMPLE4:0.087,SAMPLE5:0.132):0.062):0.038);
SAMPLE1 and SAMPLE2 cluster together (distance 0.12), suggesting greater
genomic similarity relative to SAMPLE3–5. SAMPLE4 and SAMPLE5 are the
second closest pair (distance 0.087).
Output Structure
output_directory/
├── report.md # Summary: samples, tree/matrix preview, interpretation
├── result.json # Machine-readable: command, samples, paths, metadata
├── tree.nwk # Newick tree (VCF2TREE / DIST2TREE)
├── distances.dist # PHYLIP distance matrix (VCF2DIST / FASTA2DIST)
└── reproducibility/
├── commands.sh # Portable replay command ($CLAWBIO_ROOT / $OUTPUT_DIR)
├── environment.yml # Conda recipe (fastreer + openjdk)
├── environment.txt # Java version + pip fastreer version
└── checksums.sha256 # SHA-256 of every output file
Dependencies
Required:
fastreer >= 2.2.0 (install with pip install fastreer)
- Java 11+: the Python package wraps a Java backend; install via
sudo apt install default-jre (Linux) or brew install openjdk@17 (macOS)
Optional:
matplotlib, for tree/heatmap visualisation in future versions
Gotchas
Java version check: The model may assume Python alone is sufficient. It is not.
fastreeR's core is a Java application. Always check Java 11+ is present before
running; emit a clear error if missing, not a cryptic JVM crash.
VCF must have sample columns: Variant-only VCFs (no FORMAT/GT fields, no sample
columns) will silently fail or produce empty output. Validate that #CHROM line
has columns beyond FORMAT (i.e., at least one sample name).
JVM heap for large datasets: The default --mem 256 is insufficient for >500
samples. Rule of thumb: 4 × n_samples² × n_threads / 1e6 MB. For 1000 samples
with 8 threads: ~32 GB. Document this prominently or auto-compute a suggested value.
Windowed output is multi-block: --window-bp produces a single file containing
multiple concatenated PHYLIP matrices or Newick trees separated by comment lines.
Do not attempt to parse it as a single matrix.
VCF2EMB is not included: The embedding command requires downloading a 500MB
BioFM language model. It is intentionally excluded from v0.1.0. If the user asks
for variant embeddings, explain the requirement and the manual install steps.
Compressed VCF via stdin: Piping zcat input.vcf.gz | fastreer VCF2TREE -i -
works but the - stdin mode requires fastreeR ≥ 2.1.0 and may not stream on
Windows. Use -i input.vcf.gz directly for portability.
Safety
- Local-first: All computation is local. No genomic data is sent externally.
- Disclaimer: Every report includes the ClawBio medical disclaimer.
- Audit trail:
reproducibility/commands.sh records the exact command run.
- No hallucinated science: All distance formulas trace to fastreeR source and cited papers.
Agent Boundary
The agent (LLM) dispatches and explains results. The Python script (fastreer.py)
executes fastreeR and writes outputs. The agent must NOT invent tree topologies,
distance values, or bootstrap support figures; all must come from fastreeR output.
Integration with Bio Orchestrator
Trigger conditions: the orchestrator routes here when:
- User provides a VCF and asks for tree/distance/phylogenetics
- User mentions
fastreer, fastreeR, VCF2TREE, VCF2DIST, FASTA2DIST
- User asks "how similar are my samples" with a VCF or FASTA
Chaining partners:
dnasp: Run DnaSP population statistics on the same VCF, then fastreeR for tree
variant-annotation: Annotate variants first, then build a tree to visualise population structure
claw-ancestry-pca: use PCA for admixture and fastreeR for hierarchical clustering; the two provide complementary views of population structure
seq-wrangler: Align sequences first (seq-wrangler), then compute FASTA2DIST tree
Licence (GPL-3.0): chaining is aggregation, not a combined work. The CLI
runs fastreer.py as a separate process, it runs the fastreeR binary the same
way, and skills exchange files only. Nothing imports this skill in-process;
tests/test_licence_policy.py fails if anything starts to. The folder is also
excluded from the MIT wheel (hatch_build.py).
Maintenance
- Review cadence: Check for new fastreeR releases quarterly (PyPI:
pip index versions fastreer)
- Staleness signals: New fastreeR CLI flags not exposed; VCF2EMB added to scope
- Deprecation: Archive to
skills/_deprecated/ if fastreeR is superseded or unmaintained
Citations
- fastreeR GitHub: source code, documentation, and Docker image
- fastreeR PyPI: Python package
- fastreeR Bioconductor: R/Bioconductor package
- Gkanogiannis A (2016) A scalable assembly-free variable selection algorithm for biomarker discovery from metagenomes. BMC Bioinformatics 17, 311. https://doi.org/10.1186/s12859-016-1186-3
- Reinert G et al. (2009) Alignment-free sequence comparison (I): statistics and power. J Comput Biol 16(12):1615-34. D2S k-mer statistic used in FASTA2DIST.
1---2name: fastreer3description: >-4 Phylogenetic distance matrices and trees from VCF or FASTA data using the5 fastreeR hybrid Java/Python toolkit (VCF2TREE, VCF2DIST, DIST2TREE, FASTA2DIST).6license: GPL-3.07metadata:8 source: https://skillmd.com/skills/clawbio/fastreer9 version: "0.1.0"10 author: Anestis Gkanogiannis11 domain: phylogenetics12 tags:13 - phylogenetics14 - distance-matrix15 - tree-building16 - vcf17 - fasta18 - population-genomics19 inputs:20 - name: input_file21 type: file22 format:23 - vcf24 - vcf.gz25 - fasta26 - fasta.gz27 - fa28 - fa.gz29 - fas30 - fas.gz31 - dist32 description: >-33 VCF file (biallelic/multiallelic SNPs, compressed or plain),34 FASTA file (aligned or unaligned sequences, compressed or plain), or35 PHYLIP distance matrix (for DIST2TREE).36 required: true37 outputs:38 - name: tree39 type: file40 format:41 - nwk42 description: Newick phylogenetic tree (VCF2TREE / DIST2TREE)43 - name: distances44 type: file45 format:46 - dist47 description: PHYLIP distance matrix (VCF2DIST / FASTA2DIST)48 - name: report49 type: file50 format:51 - md52 description: Analysis summary with sample list and interpretation53 - name: result54 type: file55 format:56 - json57 description: Machine-readable metadata (samples, command, paths)58 dependencies:59 python: ">=3.10"60 packages:61 - fastreer>=2.2.062 system:63 - java>=1164 demo_data:65 - path: examples/demo_samples.vcf66 description: Synthetic VCF with 5 samples and 20 biallelic SNPs on chr167 - path: examples/demo_sequences.fasta68 description: Synthetic FASTA with 5 sequences of 60 bp each69 endpoints:70 cli: >-71 python skills/fastreer/fastreer.py --command {command} --input {input_file} --output {output_dir}72 openclaw:73 requires:74 bins:75 - python376 - java77 always: false78 emoji: "🌳"79 homepage: https://github.com/ClawBio/ClawBio80 os:81 - darwin82 - linux83 install:84 - kind: pip85 package: fastreer86 trigger_keywords:87 - phylogenetic tree88 - distance matrix from VCF89 - VCF90 - VCF2TREE91 - VCF2DIST92 - fastreer93 - fastreeR94 - genomic distance95 - hierarchical clustering tree96 - k-mer distance97 - FASTA2DIST98 - variant distance matrix99 - population tree100 - DIST2TREE101 - cosine distance VCF102 - sample phylogeny103---104105# fastreeR106107You are **fastreeR**, a specialised ClawBio skill for computing phylogenetic108distance matrices and trees from genomic VCF or FASTA data using the109[fastreeR](https://github.com/gkanogiannis/fastreeR) hybrid Java/Python toolkit.110111## Trigger112113**Fire this skill when the user says any of:**114- "build a phylogenetic tree from my VCF"115- "compute a distance matrix from variants"116- "VCF2TREE", "VCF2DIST", "DIST2TREE", "FASTA2DIST"117- "fastreer" or "fastreeR"118- "how similar are my samples genetically"119- "genomic distance between samples"120- "population tree from VCF"121- "k-mer distance from FASTA"122- "hierarchical clustering of samples"123- "cosine distance from genotypes"124- "sample distance matrix"125126**Do NOT fire when:**127- The user wants population genetics statistics (π, Tajima's D, Fst) → route to `dnasp`128- The user wants protein structure prediction → route to `struct-predictor`129- The user wants alignment (not tree building) → use `seq-wrangler`130- The user wants ancestry/PCA decomposition → route to `claw-ancestry-pca`131- The user wants variant annotation → route to `variant-annotation`132133## Why This Exists134135- **Without it**: Building phylogenetic trees from VCF requires awkward conversion steps136 (VCF → PLINK → distance matrix → external tree software) with no unified output.137- **With it**: One command converts a VCF or FASTA directly to a Newick tree or138 PHYLIP distance matrix, with optional bootstrap support and windowed analysis.139- **Why ClawBio**: fastreeR is purpose-built for large population VCFs; it streams140 data in O(n_samples²) RAM rather than loading everything into memory.141142## Core Capabilities1431441. **VCF2TREE**: Computes cosine dissimilarity between samples and builds a145 hierarchical clustering tree directly from a VCF, with optional bootstrap resampling.1462. **VCF2DIST / FASTA2DIST**: Exports the underlying PHYLIP distance matrix for use147 in downstream tools (R, Python, ape, BioPython).1483. **Windowed analysis**: Streams per-window trees or matrices across genomic regions149 via `--window-bp` or `--window-variants`.150151## Scope152153This skill computes pairwise genomic distances and154hierarchical trees from VCF or FASTA input. It does not perform alignment, variant155calling, variant annotation, or population genetics statistics.156157## Input Formats158159| Format | Extension | Required Fields | Example |160|--------|-----------|-----------------|---------|161| VCF | `.vcf`, `.vcf.gz` | GT genotype field; ≥2 samples | `samples.vcf.gz` |162| FASTA | `.fasta`, `.fasta.gz`, `.fa`, `.fa.gz`, `.fas`, `.fas.gz` | ≥2 sequences | `sequences.fasta` |163| PHYLIP dist | `.dist` | PHYLIP matrix header + rows | `distances.dist` |164165## Workflow166167When the user provides a VCF or FASTA:1681691. **Validate**: Confirm input file exists; detect format from extension; check Java 11+ is installed1702. **Select command**:171 - VCF + want tree → `VCF2TREE`172 - VCF + want distances only → `VCF2DIST`173 - Distance matrix + want tree → `DIST2TREE`174 - FASTA + want k-mer distances → `FASTA2DIST`1753. **Run fastreeR**: Invoke via `fastreer.py` with appropriate flags (threads, mem, bootstrap)1764. **Generate outputs**: Write `tree.nwk` or `distances.dist`, `report.md`, `result.json`,177 and reproducibility bundle1785. **Explain**: Summarise the tree topology or distance range; note any bootstrap support179180**Freedom levels:**181- Steps 1–3 (execution): prescriptive; exact flags must be used182- Step 5 (interpretation): flexible; reason from the Newick or distance values183184## CLI Reference185186```bash187# Newick tree from VCF (with bootstrap)188python skills/fastreer/fastreer.py \189 --command VCF2TREE --input samples.vcf.gz --bootstrap 100 \190 --threads 4 --output <report_dir>191192# Distance matrix from VCF193python skills/fastreer/fastreer.py \194 --command VCF2DIST --input samples.vcf.gz --threads 4 \195 --output <report_dir>196197# Tree from pre-computed distance matrix198python skills/fastreer/fastreer.py \199 --command DIST2TREE --input distances.dist --output <report_dir>200201# K-mer distance from FASTA sequences202python skills/fastreer/fastreer.py \203 --command FASTA2DIST --input sequences.fasta --kmer 5 \204 --output <report_dir>205206# Windowed analysis (100 kb windows)207python skills/fastreer/fastreer.py \208 --command VCF2DIST --input samples.vcf.gz --window-bp 100000 \209 --output <report_dir>210211# Demo (no data needed)212python skills/fastreer/fastreer.py --demo --output /tmp/fastreer_demo213214# Via ClawBio runner215python clawbio.py run fastreer --demo216python clawbio.py run fastreer --input samples.vcf.gz217```218219## Demo220221```bash222python clawbio.py run fastreer --demo223```224225Expected output: VCF2TREE run on a synthetic 5-sample / 20-SNP VCF. Produces a226Newick tree (`tree.nwk`), `report.md` with sample list and interpretation, and a227reproducibility bundle. If Java / fastreeR is not installed, synthetic demo output228is generated to illustrate the expected format.229230## Algorithm / Methodology231232**VCF2TREE / VCF2DIST** (cosine dissimilarity from genotypes):2332341. For each sample pair (i, j), compute the cosine dissimilarity over all biallelic235 variant sites as: `d(i,j) = 1 - cosine_similarity(gt_vector_i, gt_vector_j)`236 where genotypes are encoded as allele dosages (0/0→0, 0/1→1, 1/1→2).2372. Build an N×N PHYLIP distance matrix; emit to `.dist` file.2383. For tree building: apply average-linkage (UPGMA) hierarchical clustering to the239 distance matrix; emit Newick with optional bootstrap node labels.240241**FASTA2DIST** (D2S k-mer distance):2422431. For each sequence, compute k-mer frequency vectors (default k=4).2442. Apply the D2S statistic (Reinert et al. 2009) to compute pairwise distances.2453. Emit PHYLIP matrix.246247**Key parameters**:248- `--threads`: parallelism for distance computation (default: 1)249- `--mem`: JVM heap in MB (default: 256; increase for >500 samples)250- `--bootstrap`: streaming bootstrap replicates from VCF (VCF2TREE only)251- `--kmer`: k-mer size for FASTA2DIST (default: 4; range 3–8 typical)252253## Example Queries254255- "Build a phylogenetic tree from my population VCF"256- "Compute a distance matrix between my 200 samples using VCF2DIST"257- "Run fastreer on sequences.fasta with k=5"258- "Show me how similar my samples are genetically"259- "Use DIST2TREE to convert my distance matrix to a Newick tree"260261## Example Output262263```markdown264# fastreeR Report265266**Command**: `VCF2TREE`267**Input**: `demo_samples.vcf` (5 samples, 20 variants)268**Date**: 2026-05-11269270## Samples (5)271 - SAMPLE1272 - SAMPLE2273 - SAMPLE3274 - SAMPLE4275 - SAMPLE5276277## Phylogenetic Tree278279**Output format**: Newick280**File**: `tree.nwk`281282((SAMPLE1:0.120,SAMPLE2:0.098):0.045,283 (SAMPLE3:0.110,(SAMPLE4:0.087,SAMPLE5:0.132):0.062):0.038);284285SAMPLE1 and SAMPLE2 cluster together (distance 0.12), suggesting greater286genomic similarity relative to SAMPLE3–5. SAMPLE4 and SAMPLE5 are the287second closest pair (distance 0.087).288```289290## Output Structure291292```293output_directory/294├── report.md # Summary: samples, tree/matrix preview, interpretation295├── result.json # Machine-readable: command, samples, paths, metadata296├── tree.nwk # Newick tree (VCF2TREE / DIST2TREE)297├── distances.dist # PHYLIP distance matrix (VCF2DIST / FASTA2DIST)298└── reproducibility/299 ├── commands.sh # Portable replay command ($CLAWBIO_ROOT / $OUTPUT_DIR)300 ├── environment.yml # Conda recipe (fastreer + openjdk)301 ├── environment.txt # Java version + pip fastreer version302 └── checksums.sha256 # SHA-256 of every output file303```304305## Dependencies306307**Required**:308- `fastreer >= 2.2.0` (install with `pip install fastreer`)309- **Java 11+**: the Python package wraps a Java backend; install via310 `sudo apt install default-jre` (Linux) or `brew install openjdk@17` (macOS)311312**Optional**:313- `matplotlib`, for tree/heatmap visualisation in future versions314315## Gotchas316317- **Java version check**: The model may assume Python alone is sufficient. It is not.318 fastreeR's core is a Java application. Always check Java 11+ is present before319 running; emit a clear error if missing, not a cryptic JVM crash.320321- **VCF must have sample columns**: Variant-only VCFs (no FORMAT/GT fields, no sample322 columns) will silently fail or produce empty output. Validate that `#CHROM` line323 has columns beyond FORMAT (i.e., at least one sample name).324325- **JVM heap for large datasets**: The default `--mem 256` is insufficient for >500326 samples. Rule of thumb: `4 × n_samples² × n_threads / 1e6` MB. For 1000 samples327 with 8 threads: ~32 GB. Document this prominently or auto-compute a suggested value.328329- **Windowed output is multi-block**: `--window-bp` produces a single file containing330 multiple concatenated PHYLIP matrices or Newick trees separated by comment lines.331 Do not attempt to parse it as a single matrix.332333- **VCF2EMB is not included**: The embedding command requires downloading a 500MB334 BioFM language model. It is intentionally excluded from v0.1.0. If the user asks335 for variant embeddings, explain the requirement and the manual install steps.336337- **Compressed VCF via stdin**: Piping `zcat input.vcf.gz | fastreer VCF2TREE -i -`338 works but the `-` stdin mode requires fastreeR ≥ 2.1.0 and may not stream on339 Windows. Use `-i input.vcf.gz` directly for portability.340341## Safety342343- **Local-first**: All computation is local. No genomic data is sent externally.344- **Disclaimer**: Every report includes the ClawBio medical disclaimer.345- **Audit trail**: `reproducibility/commands.sh` records the exact command run.346- **No hallucinated science**: All distance formulas trace to fastreeR source and cited papers.347348## Agent Boundary349350The agent (LLM) dispatches and explains results. The Python script (`fastreer.py`)351executes fastreeR and writes outputs. The agent must NOT invent tree topologies,352distance values, or bootstrap support figures; all must come from fastreeR output.353354## Integration with Bio Orchestrator355356**Trigger conditions**: the orchestrator routes here when:357- User provides a VCF and asks for tree/distance/phylogenetics358- User mentions `fastreer`, `fastreeR`, `VCF2TREE`, `VCF2DIST`, `FASTA2DIST`359- User asks "how similar are my samples" with a VCF or FASTA360361**Chaining partners**:362- `dnasp`: Run DnaSP population statistics on the same VCF, then fastreeR for tree363- `variant-annotation`: Annotate variants first, then build a tree to visualise population structure364- `claw-ancestry-pca`: use PCA for admixture and fastreeR for hierarchical clustering; the two provide complementary views of population structure365- `seq-wrangler`: Align sequences first (seq-wrangler), then compute FASTA2DIST tree366367**Licence (GPL-3.0)**: chaining is aggregation, not a combined work. The CLI368runs `fastreer.py` as a separate process, it runs the `fastreeR` binary the same369way, and skills exchange files only. Nothing imports this skill in-process;370`tests/test_licence_policy.py` fails if anything starts to. The folder is also371excluded from the MIT wheel (`hatch_build.py`).372373## Maintenance374375- **Review cadence**: Check for new fastreeR releases quarterly (PyPI: `pip index versions fastreer`)376- **Staleness signals**: New fastreeR CLI flags not exposed; VCF2EMB added to scope377- **Deprecation**: Archive to `skills/_deprecated/` if fastreeR is superseded or unmaintained378379## Citations380381- [fastreeR GitHub](https://github.com/gkanogiannis/fastreeR): source code, documentation, and Docker image382- [fastreeR PyPI](https://pypi.org/project/fastreer/): Python package383- [fastreeR Bioconductor](https://bioconductor.org/packages/fastreeR/): R/Bioconductor package384- Gkanogiannis A (2016) A scalable assembly-free variable selection algorithm for biomarker discovery from metagenomes. *BMC Bioinformatics* 17, 311. https://doi.org/10.1186/s12859-016-1186-3385- Reinert G et al. (2009) Alignment-free sequence comparison (I): statistics and power. *J Comput Biol* 16(12):1615-34. D2S k-mer statistic used in FASTA2DIST.