Use medaka to correct errors in genome assemblies and call haploid variants from Oxford Nanopore reads. Select a matching neural network model, process existing alignments or raw reads, and filter variant results by quality.
Polish assemblies and call variants from Oxford Nanopore data using medaka. Uses neural networks trained on specific basecaller versions. Use when improving ONT-only assemblies or calling variants from Nanopore data without short-read polishing.
Run score Not tested
See how to installReview the instructions the agent will receive before you install.
Reference examples tested with: bcftools 1.19+, minimap2 2.26+, samtools 1.19+
Before using code patterns, verify installed versions match. If versions differ:
<tool> --version then <tool> --help to confirm flagsIf code throws ImportError, AttributeError, or TypeError, introspect the installed package and adapt the example to match the actual API rather than retrying.
"Polish my ONT assembly with medaka" → Use neural networks trained on specific basecaller models to correct assembly errors and call variants from Nanopore data.
medaka_polisher -i reads.fq -d draft.fa -o polished.fa -m r1041_e82_400bps_sup_v5.0.0# Polish assembly with medaka
medaka_consensus -i reads.fastq.gz \
-d draft_assembly.fa \
-o medaka_output \
-t 4 \
-m r1041_e82_400bps_sup_v5.0.0
# Call variants against reference
medaka_variant \
-i reads.fastq.gz \
-r reference.fa \
-o output_dir \
-m r1041_e82_400bps_sup_v5.0.0
Note: Diploid variant calling has been deprecated in medaka v2.0. For diploid samples, use Clair3 instead.
Goal: Polish an ONT assembly or call variants using medaka's neural network models with explicit control over each step.
Approach: Align reads with minimap2, run medaka neural network inference on the alignment, then generate either a polished consensus or variant calls from the probability output.
# 1. Align reads to reference/draft
minimap2 -ax map-ont reference.fa reads.fastq.gz | \
samtools sort -o aligned.bam
samtools index aligned.bam
# 2. Run neural network inference
medaka inference aligned.bam consensus.hdf \
--model r1041_e82_400bps_sup_v5.0.0 \
--threads 2 # >2 threads has poor scaling
# 3. Create consensus sequence from probabilities
medaka sequence consensus.hdf reference.fa polished.fa
# 4. Call variants from probabilities
medaka vcf reference.fa consensus.hdf variants.vcf
# See all available models
medaka tools list_models
# Models are named:
# r{pore}_{chemistry}_{speed}bps_{accuracy}_{version}
# e.g., r1041_e82_400bps_sup_v5.0.0
| Model | Description |
|---|---|
| r1041_e82_400bps_sup_v5.0.0 | R10.4.1, E8.2, SUP basecalling |
| r1041_e82_400bps_hac_v5.0.0 | R10.4.1, E8.2, HAC basecalling |
| r941_min_sup_g507 | R9.4.1, MinION, SUP |
| r941_min_hac_g507 | R9.4.1, MinION, HAC |
# Check which basecaller was used in your data
# Then select matching model
# For Guppy/Dorado SUP basecalling on R10.4.1
medaka_consensus -m r1041_e82_400bps_sup_v5.0.0 ...
# For HAC basecalling
medaka_consensus -m r1041_e82_400bps_hac_v5.0.0 ...
# Polish specific region
medaka inference aligned.bam consensus.hdf \
--model r1041_e82_400bps_sup_v5.0.0 \
--region chr1:1000000-2000000
# First round
medaka_consensus -i reads.fastq.gz -d draft.fa -o round1 -m model
# Second round (diminishing returns, usually not needed)
medaka_consensus -i reads.fastq.gz -d round1/consensus.fasta -o round2 -m model
# If you already have aligned BAM
medaka inference aligned.bam consensus.hdf --model r1041_e82_400bps_sup_v5.0.0
medaka vcf reference.fa consensus.hdf variants.vcf
# Filter by quality
bcftools filter -i 'QUAL>20' variants.vcf > variants.filtered.vcf
# Get high-confidence calls
bcftools view -i 'FILTER="PASS"' variants.vcf > variants.pass.vcf
| File | Description |
|---|---|
| consensus.fasta | Polished sequence |
| consensus.hdf | Neural network outputs |
| variants.vcf | Variant calls |
| calls_to_draft.bam | Alignments used |
| Parameter | Description |
|---|---|
| -i | Input reads (FASTQ) |
| -d | Draft assembly/reference |
| -o | Output directory |
| -m | Model name |
| -t | Threads |
| -b | Batch size (GPU memory) |
| --region | Specific region to process |
# Enable GPU (if available)
medaka_consensus -i reads.fastq.gz -d draft.fa -o output \
-m r1041_e82_400bps_sup_v5.0.0 \
-b 100 \ # Increase batch size for GPU
-t 4
No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Support protein design with cloud laboratory testing for binding, expression, thermal stability, and enzyme activity. Covers sequence optimization, submitting experiments through the Adaptyv API, checking progress, and downloading experimental results.
Build flexible daily plans that account for ADHD-related difficulties with time estimation and getting started. Narrow priorities to three tasks, add transition buffers, and choose small next actions that match the energy available that day.
Build Python workflows for data recorded over time, including classification, forecasting, anomaly detection, and pattern search. Using aeon, the skill guides preprocessing, algorithm selection, feature extraction, and evaluation through interfaces compatible with scikit-learn.
Use a browser to research topics, extract page information, fill forms, and test websites. Guides agents through identifying page controls, interacting with web apps, saving screenshots or PDFs, and reusing saved login sessions.
Combine health measurements, exercise, sleep, nutrition, and mental health records to examine trends, unusual changes, and estimated risks. The skill produces personalized suggestions and text or interactive HTML reports using local data without external API calls.
Predict structures for designed protein sequences and protein complexes with AlphaFold2 or ColabFold. Inspect structure files and confidence scores to assess folding and chain interfaces, with guidance for local runs, Modal execution, and common prediction errors.