Sequencing Formula: Calculating Coverage, Depth, and Reagent Needs

Sequencing coverage and depth calculations determine whether an experiment will produce enough data to answer the biological question at hand. This article explains the core formulas used to plan sequencing experiments, estimate reagent requirements, and evaluate whether generated data meets quality thresholds. The target reader is a laboratory student, technician, researcher, or diagnostic professional who needs to calculate coverage, depth, and read counts before submitting samples for sequencing or after receiving raw data.

At a Glance

The table below summarizes the key formulas and their practical applications in sequencing experiment planning.

Understanding Coverage and Depth

Coverage and depth are related but distinct concepts in sequencing. Depth, also called depth of coverage or read depth, refers to the average number of times a nucleotide position is read during sequencing. Breadth of coverage refers to the percentage of target bases that achieve at least a specified depth threshold. Both metrics influence variant calling accuracy and the confidence placed in downstream results.

Coverage quality directly affects variant calling accuracy and final diagnosis in clinical contexts. Tools designed for clinical coverage assessment calculate depth across loci with known pathogenic variants to determine whether a sequencing run has adequately interrogated medically relevant positions. The eXNVerify tool, for example, calculates clinical depth coverage to evaluate loci with pathogenic germline and somatic variants reported in ClinVar, providing visualization options for user-defined genes of interest. This approach helps diagnosticians assess whether genetic samples have been sequenced sufficiently for reliable interpretation.

The Genome Analysis Toolkit (GATK) was developed specifically to handle the massive data sets generated by next-generation sequencing projects. The 1000 Genomes pilot alone generated nearly five terabases of data, which made writing efficient analysis tools difficult for computationally sophisticated individuals. GATK provides a structured programming framework that includes coverage calculators and single nucleotide polymorphism calling tools, enabling developers and analysts to write efficient NGS tools for large-scale projects.

Core Formulas for Sequencing Planning

Depth of Coverage Formula

The fundamental formula for calculating depth of coverage is:

Depth = (Read length × Number of reads) ÷ Target genome size

This formula assumes uniform coverage across the target region. In practice, coverage varies across the genome due to GC content, repetitive regions, and sequencing platform biases. The formula provides an average depth, not a guarantee of uniform coverage.

For example, consider a bacterial genome of 5 Mb sequenced with 150 bp paired-end reads. If the sequencing run produces 2 million read pairs, the total bases generated are 150 × 2 × 2,000,000 = 600 Mb. Dividing by the 5 Mb genome gives an average depth of 120×.

Required Reads Formula

To determine how many reads are needed for a desired depth, rearrange the depth formula:

Number of reads = (Desired depth × Target genome size) ÷ Read length

For a human whole genome of approximately 3 Gb at 30× depth with 150 bp reads:

Number of reads = (30 × 3,000,000,000) ÷ 150 = 600,000,000 reads

This calculation assumes perfect efficiency. Real sequencing runs have losses from quality filtering, adapter contamination, and duplicate reads. Adding 10 to 20 percent excess reads is a common planning practice.

Coverage Breadth Calculation

Breadth of coverage is calculated as:

Breadth = (Number of bases covered at or above threshold) ÷ Total target bases

This metric is often reported at specific depth thresholds such as 10×, 20×, or 30×. For targeted panels, breadth at the minimum depth required for confident variant calling is the critical quality metric.

The mosdepth tool provides rapid calculation of genome-wide sequencing coverage from BAM or CRAM files. It measures depth at each nucleotide position or for sets of genomic regions specified as BED files for capture regions or fixed-size windows for copy-number calling. Mosdepth produces coverage summaries faster than existing tools and offers flexibility in the types of coverage profiles generated.

Sequencing Types and Their Coverage Requirements

Whole Genome Sequencing

Whole genome sequencing aims to cover the entire genome of an organism. Coverage requirements depend on the application. High coverage sequencing of known genes detects monogenic variants, while low coverage whole genome sequencing can support polygenic risk score calculations when combined with imputation.

Low coverage whole genome sequencing at 0.5× achieved imputation accuracy r² values greater than 0.90 for ten samples including those of African and Ashkenazi Jewish ancestry. Genome-wide polygenic scores calculated using low coverage whole genome sequencing and genotyping array followed by imputation were highly correlated for coronary artery disease, breast cancer, and atrial fibrillation, with r² values from 0.93 to 0.97.

For underrepresented populations, low coverage sequencing at a depth of 4× or greater captures variants of all frequencies more accurately than commonly used GWAS arrays at comparable cost. Lower depths of 0.5 to 1× performed comparably to low-density GWAS arrays. Four-fold sequencing detects 45 percent of singletons and 95 percent of common variants identified in high coverage African whole genomes.

Low Coverage Sequencing Applications

Low coverage whole genome sequencing presents a cost-effective solution for genotyping in applications requiring high marker density and reduced costs. In eggplant genotyping, sequencing coverages as low as 1× and 2× achieved high accuracy but lacked sufficient sensitivity and genotypic concordance. Three-fold sequencing reached approximately 10 percent less sensitivity than 5× while maintaining genotypic concordance above 90 percent at any depth of coverage threshold.

Portable Nanopore sequencing at low coverage has demonstrated potential for agricultural applications. At sequencing coverages between 2× and 4×, the correlation between Oxford Nanopore Technologies breeding values and SNP array based breeding values was greater than 0.92 when imputation was used and greater than 0.88 without imputation. At an average sequencing coverage of 0.5×, the correlation was between 0.85 and 0.92 using imputation depending on the trait.

Ultra low depth whole genome sequencing has been used for preimplantation kinship identification. A model developed using data from 789 biopsied embryos at an average depth of 0.02× predicted embryonic relationships via a heterozygosity based relationship ratio. At optimized thresholds of 800,000 unique reads and 800 co-detected SNPs, the model achieved 100 percent accuracy in discriminating full sibling, unrelated, and identical relationships.

Targeted Sequencing Panels

Targeted panels focus sequencing on specific genomic regions. The target size is the sum of all captured regions instead of the whole genome. Coverage calculations use the panel size in the denominator.

For a panel targeting 500 genes with a total size of 2 Mb, achieving 100× depth with 150 bp reads requires:

Number of reads = (100 × 2,000,000) ÷ 150 = 1,333,333 reads

This calculation assumes uniform capture efficiency. In practice, capture panels show variable enrichment across regions, and some regions may require additional optimization.

Exome Sequencing

Exome sequencing targets the protein coding regions of the genome, approximately 30 Mb in humans. The smaller target size means lower read requirements compared to whole genome sequencing for equivalent depth.

For exome sequencing at 100× depth with 150 bp reads:

Number of reads = (100 × 30,000,000) ÷ 150 = 20,000,000 reads

Exome sequencing projects require software pipelines for variant discovery. A software pipeline for the discovery of variations in exome sequencing projects was described in 2012, addressing the specific challenges of analyzing exome capture data.

Metagenomic Sequencing

Metagenomic sequencing presents unique coverage considerations because the target is a community of organisms instead of a single genome. Coverage calculations must account for the relative abundance of different species and the diversity of the community.

Fragmentation and coverage variation in viral metagenome assemblies affect diversity calculations. The variable coverage across viral genomes in a metagenome can lead to incomplete assemblies and biased diversity estimates.

Taxonomic classification performance depends on reference database selection and confidence score settings. Higher confidence scores, which require greater k-mer agreement, generally decrease classification rates. For larger databases such as Standard, nt, and GTDB r202, precision and F1 scores improved significantly with increasing confidence scores. A comprehensive reference database combined with a moderate confidence score of 0.2 or 0.4 significantly improves classification accuracy and sensitivity.

Reagent and Cost Calculations

Per Sample Reagent Estimation

Reagent needs depend on the sequencing platform, library preparation method, and desired coverage. The core calculation is:

Reagent cost per sample = Library preparation cost + (Sequencing cost per run ÷ Samples per run)

For pooled sequencing, the number of samples that can be multiplexed in a single run depends on the reads required per sample and the total reads the platform produces:

Samples per run = Total reads per run ÷ Reads required per sample

For a platform producing 400 million reads per run and a sample requiring 12.5 million reads for 30× human whole genome coverage:

Samples per run = 400,000,000 ÷ 12,500,000 = 32 samples

Pooling Calculations

When pooling samples with different coverage requirements, calculate the volume of each library to combine based on molar concentration and desired read proportion.

The proportion of reads allocated to each sample in a pool is:

Proportion = Desired reads for sample ÷ Total desired reads for all samples

For a pool of 48 samples where each requires equal coverage, each sample receives 1/48 of the total reads. If the platform produces 400 million reads, each sample receives approximately 8.3 million reads.

Cost Comparison Across Approaches

Low coverage sequencing approaches can reduce costs while maintaining data quality. Low coverage sequencing at 4× captures variants more accurately than all commonly used GWAS arrays investigated at comparable cost. This approach surmounts the problems induced by the ascertainment of common genotyping arrays and effectively identifies novel variation particularly in underrepresented populations.

Genotyping by sequencing has become popular due to lower cost and greater genome coverage including structural variants compared to SNP arrays. Portable sequencing devices combine the benefits of genotyping by sequencing with portability and decreased turnaround time, introducing potential for in-house clinical genetic disease risk screening or calculating genomic breeding values on farm in agriculture.

Tools for Coverage Calculation

Command Line Tools

Several command line tools calculate coverage from sequencing alignment files. Mosdepth measures depth from BAM or CRAM files at each nucleotide position or for sets of genomic regions. It accepts BED files to evaluate coverage across capture regions or fixed size windows for copy number calling. Mosdepth uses a simple algorithm that is computationally efficient and produces coverage summaries faster than existing tools.

PanDepth is an ultrafast and efficient tool for calculating coverage and depth from sequencing alignments. It outperforms other tools in computation time and memory efficiency for both BAM and CRAM format alignment files regardless of read length. PanDepth employs chromosome parallel computation and optimized data structures, accepts sorted or unsorted BAM and CRAM files as well as GTF, GFF, and BED formatted interval files or a specific window size. When provided with a reference genome sequence and the option to enable GC content calculation, PanDepth includes GC content statistics that enhance the accuracy and reliability of copy number variation analysis.

Clinical Coverage Tools

Clinical applications require tools that evaluate coverage at medically relevant loci. The eXNVerify tool calculates clinical depth coverage, a measure that evaluates loci with pathogenic germline and somatic variants reported in ClinVar. The tool provides visualization options for user defined genes of interest and has been applied to BRCA1, TP53, and CFTR. It runs in a Docker container, which allows users who are not proficient in complex software to perform coverage analysis without special technical preparation.

Choosing a Coverage Tool

Select a coverage tool based on the specific requirements of the experiment. Consider the input file format, the need for interval based analysis, the size of the dataset, and the computational resources available. Tools that support parallel computation reduce processing time for large datasets. Tools that accept BED files allow evaluation of coverage across capture regions or user defined genomic intervals.

Practical Workflow for Calculating Sequencing Requirements

Step 1: Define the Biological Question

Determine what variants or features the experiment must detect. This decision sets the required depth and breadth of coverage. Clinical diagnostic applications require higher depth at known pathogenic loci. Population genetic studies may use lower depth with imputation. Metagenomic studies require consideration of community diversity and abundance.

Step 2: Determine Target Size

Calculate the total size of the target region. For whole genome sequencing, use the genome size of the organism. For targeted panels, sum the sizes of all captured regions. For exome sequencing, use the total coding region size.

Step 3: Select Desired Depth

Choose the depth required for confident variant calling. Higher depth increases confidence but also increases cost. Consider the variant types of interest, the expected allele frequencies, and the error rates of the sequencing platform.

Step 4: Calculate Required Reads

Apply the required reads formula:

Number of reads = (Desired depth × Target genome size) ÷ Read length

Add 10 to 20 percent excess reads to account for quality filtering and duplicate reads.

Step 5: Determine Platform and Run Configuration

Match the required reads to the output of available sequencing platforms. Calculate the number of samples that can be pooled per run and the number of runs needed for the total sample count.

Step 6: Estimate Reagent Volumes

Calculate library preparation reagents per sample and sequencing reagents per run. Multiply by the number of samples and runs to determine total reagent requirements.

Step 7: Document the Plan

Record the target size, desired depth, read length, required reads, platform, pooling strategy, and reagent volumes. This documentation supports reproducibility and troubleshooting.

Records and Measurements

Essential Records for Sequencing Experiments

Maintain records of the following parameters for each sequencing experiment:

• Target region size and coordinates
• Desired depth and breadth thresholds
• Read length and read count generated
• Average depth achieved
• Breadth at specified depth thresholds
• Number of reads passing quality filters
• Duplicate read rate
• Reagent lot numbers and volumes used
• Platform and software versions

Quality Metrics to Track

Track the following quality metrics for each run:

• Depth distribution across target regions
• Percentage of target bases at or above minimum depth
• Uniformity of coverage
• GC bias
• Duplicate rate
• Mapping rate
• Error rate

The Laboratory Quality Management System Handbook from the World Health Organization provides guidance on establishing quality systems in laboratories. Quality management principles apply to sequencing workflows, including documentation, validation, and quality control procedures.

Common Failure Patterns in Coverage Planning

Insufficient Depth for Variant Calling

The most common failure is generating insufficient depth for confident variant calling. This occurs when the required reads calculation does not account for losses from quality filtering, adapter contamination, and duplicate reads. Low coverage regions may fall below the minimum depth threshold even when average depth appears adequate.

Uneven Coverage Across Target Regions

Coverage is rarely uniform across a genome or target region. GC rich and GC poor regions often show reduced coverage. Repetitive regions may have ambiguous read mapping. Capture panels may show variable enrichment efficiency across regions. Evaluate breadth of coverage at the minimum depth threshold instead of relying solely on average depth.

Overpooling Samples

Pooling too many samples per run reduces the reads available per sample. Calculate the reads required per sample and verify that the pooling strategy provides sufficient reads for the lowest coverage sample in the pool.

Underestimating Reagent Needs

Reagent calculations based on theoretical yields may underestimate actual needs. Library preparation has variable yields, and sequencing runs may produce fewer reads than expected. Include excess capacity in reagent planning.

Ignoring Platform Specific Biases

Different sequencing platforms have different error profiles and bias patterns. Long read platforms have different coverage characteristics than short read platforms. Account for platform specific biases when calculating coverage requirements.

Limitations of Coverage Calculations

Average Depth Does Not Guarantee Uniform Coverage

The depth formula provides an average across the target region. Actual coverage varies substantially across the genome. Some regions may have zero coverage while others have many times the average depth. Evaluate breadth of coverage at relevant depth thresholds to assess whether the data meets the needs of the experiment.

Imputation Dependent Analyses Have Specific Requirements

Low coverage sequencing with imputation requires appropriate reference panels and analysis methods. The accuracy of imputation depends on the relatedness between the study population and the reference panel. Low coverage sequencing at 0.5× achieved imputation accuracy r² values greater than 0.90 for samples including those of African and Ashkenazi Jewish ancestry, but this performance depends on the availability of suitable reference data.

Coverage Requirements Vary by Application

Different applications require different coverage levels. High coverage sequencing of known genes detects monogenic variants. Low coverage sequencing with imputation supports polygenic risk score calculations. Metagenomic studies require coverage sufficient to detect low abundance species. Match coverage to the specific requirements of the application.

Computational Resources Affect Analysis Feasibility

Coverage calculation and variant calling require substantial computational resources for large datasets. The massive data sets generated by next generation sequencing make writing efficient and robust analysis tools difficult. Tools that use parallel computation and optimized data structures reduce processing time and memory requirements.

Safety and Regulatory Context

Biosafety Considerations

Sequencing laboratories must follow biosafety guidelines for handling biological samples. The Laboratory Biosafety Manual from the World Health Organization provides guidance on safe handling of biological materials. Follow institutional biosafety protocols for sample collection, nucleic acid extraction, and sequencing library preparation.

Quality Management

The Laboratory Quality Management System Handbook from the World Health Organization describes the components of a quality management system for laboratories. Quality management principles apply to sequencing workflows, including method validation, quality control, documentation, and corrective action.

Method Validation

The Bioanalytical Method Validation Guidance from the U.S. Food and Drug Administration provides principles for validating analytical methods. While developed for bioanalytical methods, the principles of accuracy, precision, selectivity, sensitivity, and reproducibility apply to sequencing based assays used in diagnostic contexts.

Clinical Interpretation

The Assay Guidance Manual from the National Center for Advancing Translational Sciences provides guidance on assay development and validation. For clinical applications, variant interpretation requires consideration of population databases and functional evidence. The NCBI Literature Resources provide access to the biomedical literature needed for variant interpretation.

Professional Escalation Criteria

When to Escalate Coverage Problems

Escalate to a supervisor or bioinformatics specialist when:

• Average depth falls below the minimum required for the application
• Breadth of coverage at the minimum depth threshold is below acceptable levels
• Coverage is highly uneven across target regions
• Multiple samples in a pool show consistently low coverage
• Quality metrics indicate platform or reagent problems
• Variant calling results are discordant with expected findings

When to Repeat Sequencing

Repeat sequencing when:

• The proportion of target bases at minimum depth is insufficient for confident interpretation
• Critical regions have zero coverage
• Quality metrics indicate systematic bias
• Sample identity or quality is questionable

When to Seek Specialized Consultation

Consult a bioinformatics specialist when:

• Coverage calculations require complex modeling of non uniform coverage
• Imputation based analyses require specialized methods
• Metagenomic community analysis requires diversity aware coverage planning
• Clinical interpretation requires integration of multiple variant evidence types

Frequently Asked Questions

What is the difference between coverage and depth in sequencing?

Depth of coverage refers to the average number of times each nucleotide position is read during sequencing. Breadth of coverage refers to the percentage of target bases that achieve at least a specified depth threshold. Depth is calculated as total bases generated divided by target size. Breadth is calculated as the number of bases covered at or above a threshold divided by total target bases.

How do I calculate the number of reads needed for a sequencing experiment?

Multiply the desired depth by the target genome size, then divide by the read length. For a human whole genome of 3 Gb at 30× depth with 150 bp reads, the calculation is 30 × 3,000,000,000 ÷ 150 = 600,000,000 reads. Add 10 to 20 percent excess reads to account for quality filtering and duplicate reads.

What coverage depth is needed for whole genome sequencing?

Coverage requirements depend on the application. High coverage sequencing of known genes detects monogenic variants. Low coverage whole genome sequencing at 0.5× supports polygenic risk score calculations when combined with imputation. Low coverage sequencing at 4× or greater captures variants of all frequencies more accurately than commonly used GWAS arrays at comparable cost.

How does low coverage sequencing compare to genotyping arrays?

Low coverage sequencing at a depth of 4× or greater captures variants of all frequencies more accurately than all commonly used GWAS arrays investigated at comparable cost. Lower depths of 0.5 to 1× performed comparably to commonly used low density GWAS arrays. Low coverage sequencing is also sensitive to novel variation and surmounts the problems induced by the ascertainment of common genotyping arrays.

What tools can I use to calculate coverage from sequencing data?

Mosdepth measures depth from BAM or CRAM files at each nucleotide position or for sets of genomic regions specified as BED files or fixed size windows. PanDepth calculates coverage and depth from BAM and CRAM format alignment files with ultrafast computation speeds and memory efficiency. The eXNVerify tool calculates clinical depth coverage for loci with pathogenic variants reported in ClinVar.

How do I calculate how many samples can be pooled in a single sequencing run?

Divide the total reads produced by the platform per run by the reads required per sample. For a platform producing 400 million reads per run and a sample requiring 12.5 million reads for 30× human whole genome coverage, the calculation is 400,000,000 ÷ 12,500,000 = 32 samples per run.

What is the minimum coverage needed for variant calling?

The minimum coverage depends on the variant type, the expected allele frequency, and the error rate of the sequencing platform. Higher depth increases confidence but also increases cost. Evaluate breadth of coverage at the minimum depth threshold required for confident variant calling instead of relying solely on average depth.

How does coverage affect metagenomic sequencing results?

Metagenomic sequencing coverage must account for the relative abundance of different species and the diversity of the community. Fragmentation and coverage variation in viral metagenome assemblies affect diversity calculations. Taxonomic classification performance depends on reference database selection and confidence score settings, with a comprehensive reference database combined with a moderate confidence score improving classification accuracy and sensitivity.

Related Diagnostic Guides

• How to Calculate the Specific Activity of an Enzyme: Formula and Examples
• How to Calculate the Number of PCR Cycles Needed for Amplification
• How to Calculate Transformation Efficiency: Formula, Examples, and Common Pitfalls
• Metagenomic Next-Generation Sequencing (mNGS) for Veterinary Diagnostics: Challenges and Applications
• How to Calculate the Number of Molecules in a DNA Sample

References and Further Reading

• Laboratory Quality Management System Handbook. World Health Organization.
• Laboratory Biosafety Manual. World Health Organization.
• Assay Guidance Manual. National Center for Advancing Translational Sciences.
• Bioanalytical Method Validation Guidance. U.S. Food and Drug Administration.
• NCBI Literature Resources. National Center for Biotechnology Information.
• The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data.. Genome research, 2010.
• Low coverage whole genome sequencing enables accurate assessment of common variants and calculation of genome-wide polygenic scores.. Genome medicine, 2019.
• eXNVerify: coverage analysis for long and short-read sequencing data in clinical context.. F1000Research, 2022.
• PanDepth, an ultrafast and efficient genomic tool for coverage calculation.. Briefings in bioinformatics, 2024.
• Genomic prediction using low-coverage portable Nanopore sequencing.. PloS one, 2021.
• Low-coverage sequencing cost-effectively detects known and novel variation in underrepresented populations.. American journal of human genetics, 2021.
• Benchmarking of low coverage sequencing workflows for precision genotyping in eggplant.. BMC plant biology, 2025.
• Mosdepth: quick coverage calculation for genomes and exomes.. Bioinformatics (Oxford, England), 2018.
• PerturbPlan: An analytical framework for designing Perturb-seq experiments. 2026.
• Machine learning-based assessment of the healthy human gut mycobiota landscape using ITS1 DNA metabarcoding data.. 2026.
• The Effects of a Prebiotic Formula Promoting &lt,i&gt,Akkermansia muciniphila&lt,/i&gt, (AKK) on Gut Health: A Single-Centre, Randomised Controlled Trial.. 2026.
• Comparative Genome Analysis of Illumina, Nanopore, and Hybrid Approaches: A Case Study of the Aquaculture Isolate 160P.. 2026.
• Construction and validation of a preimplantation kinship identification model using ultra-low-depth whole genome sequencing.. 2026.
• Benchmarking AlphaMissense against ClinVar for Diagnostic Interpretation of Missense Variants in Inherited Retinal Diseases. Ophthalmology Science, 2025.
• Impact of database choice and confidence score on the performance of taxonomic classification using Kraken2. aBIOTECH, 2024.
• A software pipeline for the discovery of variations in exome sequencing projects. 2012.
• A high quality assembly of the Nile Tilapia (Oreochromis niloticus) genome reveals the structure of two sex determination regions. BMC Genomics, 2017.
• Estimating Global Methylation and Erasure Using Low-Coverage Whole-Genome Bisulfite Sequencing (WGBS ). Methods in Molecular Biology, 2021.
• Fragmentation and coverage variation in viral metagenome assemblies, and their effect in diversity calculations. Frontiers in Bioengineering and Biotechnology, 2015.
• Improved computations for relationship inference using low-coverage sequencing data. BMC Bioinformatics, 2023.

This article is educational and does not replace validated laboratory procedures, institutional biosafety review, manufacturer instructions, or professional interpretation.