Smrna preprocessing

Version Compatibility

Reference examples tested with: cutadapt 4.4+, fastp 0.23+, matplotlib 3.8+

Before using code patterns, verify installed versions match. If versions differ:

• Python: pip show <package> then help(module.function) to check signatures
• CLI: <tool> --version then <tool> --help to confirm flags

If code throws ImportError, AttributeError, or TypeError, introspect the installed package and adapt the example to match the actual API rather than retrying.

Small RNA Preprocessing

"Preprocess my small RNA-seq reads" → Remove 3' adapter sequences and size-select reads in the small RNA range (18-30 nt for miRNA, 24-32 nt for piRNA) before quantification or discovery.

• CLI: cutadapt -a ADAPTER -m 18 -M 30 -o trimmed.fastq input.fastq

Adapter Trimming with Cutadapt

Goal: Remove 3' adapter sequences and size-select reads in the small RNA range.

Approach: Run cutadapt with the kit-specific adapter, minimum/maximum length filters, and discard reads without adapter.

Small RNA libraries have specific 3' adapters that must be removed:

# Standard Illumina TruSeq small RNA adapter
cutadapt \
    -a TGGAATTCTCGGGTGCCAAGG \
    -m 18 \
    -M 30 \
    --discard-untrimmed \
    -o trimmed.fastq.gz \
    input.fastq.gz

# -a: 3' adapter sequence
# -m 18: Minimum length (miRNAs are 18-25 nt)
# -M 30: Maximum length (exclude longer fragments)
# --discard-untrimmed: Remove reads without adapter (likely not small RNA)

Common Small RNA Adapters

Kit	3' Adapter Sequence
Illumina TruSeq	TGGAATTCTCGGGTGCCAAGG
NEBNext	AGATCGGAAGAGCACACGTCT
QIAseq	AACTGTAGGCACCATCAAT
Lexogen	TGGAATTCTCGGGTGCCAAGGAACTCCAGTCAC

Size Selection

# Filter by length after trimming
cutadapt \
    -a TGGAATTCTCGGGTGCCAAGG \
    -m 18 -M 26 \
    -o mirna_length.fastq.gz \
    input.fastq.gz

# miRNA: 18-26 nt (typically 21-23 nt)
# piRNA: 26-32 nt
# snoRNA: variable, typically longer

Quality Trimming

# Trim low-quality bases from 3' end before adapter removal
cutadapt \
    -q 20 \
    -a TGGAATTCTCGGGTGCCAAGG \
    -m 18 \
    -o trimmed.fastq.gz \
    input.fastq.gz

Using fastp for Small RNA

# fastp with small RNA settings
fastp \
    --in1 input.fastq.gz \
    --out1 trimmed.fastq.gz \
    --adapter_sequence TGGAATTCTCGGGTGCCAAGG \
    --length_required 18 \
    --length_limit 30 \
    --html report.html

# Note: fastp auto-detects adapters but specifying is more reliable

Collapse Identical Reads

For small RNAs, collapsing identical sequences reduces computation:

# Using seqkit
seqkit rmdup -s trimmed.fastq.gz -o collapsed.fasta

# Using fastx_toolkit (legacy)
fastx_collapser -i trimmed.fastq -o collapsed.fasta

Python Preprocessing

import gzip
from collections import Counter

def collapse_reads(fastq_path):
    '''Collapse identical sequences and count occurrences'''
    counts = Counter()

    with gzip.open(fastq_path, 'rt') as f:
        while True:
            header = f.readline()
            if not header:
                break
            seq = f.readline().strip()
            f.readline()  # +
            f.readline()  # qual

            # Only keep reads in miRNA size range
            if 18 <= len(seq) <= 26:
                counts[seq] += 1

    return counts

# Write collapsed FASTA
def write_collapsed_fasta(counts, output_path):
    with open(output_path, 'w') as f:
        for i, (seq, count) in enumerate(counts.most_common()):
            f.write(f'>seq_{i}_x{count}\n{seq}\n')

QC Metrics for Small RNA

Key metrics to check:

• Read length distribution (should peak at 21-23 nt for miRNA)
• Adapter content (high if library is good)
• Percentage of reads in target size range

import matplotlib.pyplot as plt
from collections import Counter

def plot_length_distribution(fastq_path):
    lengths = Counter()
    with gzip.open(fastq_path, 'rt') as f:
        for i, line in enumerate(f):
            if i % 4 == 1:  # Sequence line
                lengths[len(line.strip())] += 1

    plt.bar(lengths.keys(), lengths.values())
    plt.xlabel('Read Length')
    plt.ylabel('Count')
    plt.title('Small RNA Length Distribution')
    plt.savefig('length_dist.png')

Related Skills

• mirdeep2-analysis - Novel miRNA discovery
• mirge3-analysis - Fast miRNA quantification
• read-qc/adapter-trimming - General adapter trimming