This protocol describes an end-to-end workflow for telomere-to-telomere sequencing of the human genome using the Oxford Nanopore PromethION platform. The protocol includes three separate sequencing experiments, using the Ultra-Long DNA Sequencing Kit V14 (SQK-ULK114), the Assembly Polishing Kit V14 (SQK-APK114), and the Pore-C protocol with the Ligation Sequencing Kit V14 (SQK-LSK114). A total of four PromethION Flow Cells are recommended for telomere-to-telomere sequencing of a single human sample.

The protocol describes each experiment individually; however, the Ultra-Long DNA Sequencing Kit and Pore-C library preparation steps will be carried out across multiple days. All optional and required pause steps will be highlighted throughout the protocol.

This protocol was developed in collaboration with the UCSC Nanopore Production Center, led by Dr. Karen Miga.

To achieve in-depth telomere-to-telomere sequencing of a sample, three different datasets must be generated to give high-accuracy data, ultra-long reads, as well as chromatin conformation capture data.

The following three experiments are set up:

• Ultra-long DNA sequencing experiment: This experiment yields ultra-long reads using the Ultra-Long DNA Sequencing Kit V14 (SQK-ULK114), generating a viscous library of ultra-long DNA fragments that requires careful handling to maintain the long fragments. The DNA extraction and library preparation processes are performed in duplicate side-by-side, taking ~3.5 hours with an overnight elution, yielding a total of ten library loads across two flow cells. This means five library loads are required per flow cell across a 140-hour sequencing run.

• Assembly Polishing Kit experiment: This experiment generates high-accuracy data using the Assembly Polishing Kit V14 (SQK-APK114). It takes approximately one day to complete the library preparation step and sequencing is performed on one PromethION Flow Cell.

• Pore-C sequencing experiment: This experiment generates chromatin conformation capture data using the Pore-C protocol and the Ligation Sequencing Kit V14 (SQK-LSK114). The Pore-C DNA extraction takes ~3 hours of hands-on time over three days with two overnight steps. The library preparation step takes ~60 minutes of hands-on time and is loaded on one PromethION Flow Cell. This experiment has been developed by Oxford Nanopore Technologies and the following published literature: Lieberman-Aiden et al., 2009; Comet et al., 2011; Belton et al., 2012; Gavrilov, Golov and Razin, 2013; Nagano et al., 2015; Belaghzal, Dekker and Gibcus, 2017; Ulahannan et al., 2019. This experiment intends to manipulate cell suspensions to capture three-dimensional interactions of DNA within chromatin. This workflow has been written using NlaIII restriction enzyme and the heat denaturation method. For further information on protocol considerations, please see the Restriction Enzyme Pore-C info sheet.

Steps in the sequencing workflow

Prepare for your experiment

You will need to:

• Ensure you have your human cell line or whole blood samples ready.
• Ensure you have your sequencing kits, the correct equipment, and third-party reagents.
• Download the MinKNOW software for acquiring and analysing your data.
• Check your flow cells to ensure they have enough pores for a good sequencing run.

Protocol workflow

The tables below are an overview of the steps required in each experiment, including timings and optional stopping points.

Ultra-long DNA experiment
Note: this experiment is performed in duplicate, each prep yielding five library loads per flow cell.

Steps	Process	Time	Stop option
Cell isolation	Isolate white blood cells from whole blood or cells from cell culture	30 minutes	4°C overnight
Ultra-long DNA extraction	Extract ultra-long high molecular weight DNA	190 minutes	Stored at room temperature overnight, store at 4°C for short-term storage
Library preparation	Tagment your DNA using a diluted fragmentation mix, attach the sequencing adapters and clean up the sample be precipitating your DNA and eluting overnight	190 minutes	Overnight elution at room temperature. 4°C short-term storage or for repeated use, such as reloading your flow cell. -80°C for single-use, long-term storage. We strongly recommend sequencing your library as soon as it is adapted. All excess adapted DNA library should be stored at 4°C or on ice until use.
Priming and loading your flow cell	Prime your flow cell and load the prepared library for sequencing	~ 30 minutes (5 minutes hands-on time)
Washing and reloading your flow cell	Wash your flow cell and reload the prepared library for further sequencing every 20-24 hours until you have sequenced five library loads on a flow cell.	~ 45 minutes (5-10 minutes hands-on time)

Assembly Polishing Kit experiment

Steps	Process	Time	Stop option
DNA extraction and shearing	Extract DNA from either whole blood cells or cell culture. Shear the extracted DNA using the Covaris g-TUBE™	225–300 minutes (hands-on time)	For either extraction option you can store the DNA at 4°C until the shearing step. Following DNA shearing, the sample can be stored at 4°C until library preparation.
Library preparation	Repair DNA and prepare the DNA ends for adapter attachment for the single cycle polymerase fill-in step. Finally, attach the sequencing adapters for the DNA ends.	175 minutes (hands-on time)	There are multiple optional pause steps at 4°C overnight. We strongly recommend sequencing your library as soon as the Rapid Adapter (RA) is attached.
Priming and loading your flow cell	Prime the flow cell and load the prepared library for sequencing.	~ 30 minutes (5 minutes hands-on time)

Pore-C experiment

Steps	Process	Time	Stop option
Cell isolation	Isolate white blood cells from whole blood or cells from cell culture	160 minutes	The custom SPRI beads can be made and stored at 4°C before use. Snap Freeze aliquots of white blood cells and store at -80°C until the experiment can begin.
Pore-C extraction	Crosslink the three-dimensional DNA interactions within the nucleus of isolated cells. Next, permeabilise the cells and denature the chromatin. Cleave the genome with a restriction enzyme and ligate the cohesive ends of proximal crosslinked monomers into chimeric Pore-C polymers held in proximity. Degrade the protein structures to release the chimeric Pore-C polymers into solution and finally, purify the Pore-C extract.	Day 1: 50 minutes hands on time 2.5 hour procedure time and overnight Day 2: 10 minutes hands on time, 6 hours procedure time and overnight Day 3: 40 minute hands on time, 1 hour 50 procedure time and optional overnight step	Aside from the multiple overnight incubations, there are a couple of optional pause steps. Snap-freeze the crosslinked aliquots in liquid nitrogen. Store frozen sample pellets at –80°C and use within one year. The extracted DNA can be stored at 4°C overnight until library preparation.
Library preparation	Repair DNA and prepare the ends for sequencing adapter attachment.	60 minutes	Overnight storage at 4°C following DNA library elution. We strongly recommend sequencing your library as soon as it is adapted.
Priming and loading your flow cell	Prime the flow cell and load the prepared library for sequencing.	~ 30 minutes (5 minutes hands-on time)

Sequencing and analysis

• For each experiment, start a sequencing run using the MinKNOW software which will collect raw data from the device. Live basecalling is performed to support output estimates during sequencing.

Further details for each experiment set-up and basecalling are outlined in the "Data acquisition and basecalling" section of each experiment:

• For the Ultra-long DNA experiment re-basecall your data after sequencing has completed using Dorado.
• For the Assembly Polishing Kit experiment re-basecall your data after sequencing has completed using Dorado, then perform read correction of your basecalled data using Herro.
• For the Pore-C experiment proceed with the data that has been is basecalled live in MinKNOW during sequencing.

Telomere-to-telomere assembly:

• Finally, perform the telomere-to-telomere assembly using your basecalled data from all three experiments following the "Downstream analysis" section of the protocol.

We recommend different sample preparations due to different input requirements for each experiment, with the option to use either human cell lines or human whole blood. Other methods are available and may be more appropriate for your lab; however, please ensure to yield enough input required for each library preparation. It is also worth noting that depending on how DNA is extracted from a sample, certain chemical contaminants may remain in the purified DNA, which can affect library preparation efficiency and sequencing quality. Read more about contaminants on the Contaminants page.

Human whole blood: approximately a total of 10–15 ml of blood is required for all sample preparation steps. The whole blood can be collected in an anticoagulant such as K2-EDTA but we do not recommend mixing with other additives as they may interfere with the Pore-C DNA extraction or the DNA sequencing run.

Human whole blood input requirements:

• Ultra-long DNA experiment: 3.2 ml
• Assembly Polishing Kit experiment: 1 ml
• Pore-C experiment: 5-10 ml

Human cell lines from culture can also be used. We recommend isolating them from cell culture using standard techniques. However, for the Assembly Polishing Kit experiment, we have included an extraction protocol from cell culture.

• Ultra-long DNA experiment: 6 million cells
• Assembly Polishing Kit experiment: 5 million cells
• Pore-C experiment: 10 million cells

We have validated and recommend the use of all the third-party reagents used in this protocol. Alternatives have not been tested by Oxford Nanopore Technologies.

For all third-party reagents, we recommend following the manufacturer's instructions to prepare the reagents for use.

Name	Acronym	Cap colour	Number of vials	Fill volume per vial (µl)
AP Adapter	APA	Blue	1	15
AP Primer	APP	Yellow	1	35
Rapid Adapter	RA	Green	1	15
Long Fragment Buffer	LFB	Clear	1	7,500
Ligation Buffer	LNB	White	1	200
AMPure XP Beads	AXP	Amber	3	1,200
Adapter Buffer	ADB	Clear	1	100
Elution Buffer	EB	Black	1	500
Sequencing Buffer	SB	Red	1	700
Library Beads	LIB	Pink	1	600
Library Solution	LIS	White cap, pink label	1	600
Flow Cell Flush	FCF	Clear cap, light blue label	1	8,000
Flow Cell Tether UL	FTU	Purple	1	600
AP Mix	APM	Clear cap	1	60

Name	Acronym	Cap colour	Number of vials	Fill volume per vial (µl)
Rapid Adapter	RA	Green	1	40
Fragmentation Mix	FRA	Amber	1	50
FRA Dilution Buffer	FDB	Clear	1	1,600
Elution Buffer	EB	Black	2	1,500
Extraction EB	EEB	Orange	3	1,700
Sequencing Buffer UL	SBU	Red	2	1,000
Loading Solution UL	LSU	White cap, pink label	1	200
Flush Tether UL	FTU	Purple	1	600
Flow Cell Flush	FCF	Blue	2	15,500
Precipitation Buffer	PTB	Blue	2	1,700
Precipitation Star	PS	Yellow	6	1 star

Name	Acronym	Cap colour	Number of vials	Fill volume per vial (µl)
Elution Buffer	EB	Black	1	1,500
Sequencing Buffer UL	SBU	Red	2	1,000
Loading Solution UL	LSU	White cap, pink label	1	200
Flush Tether UL	FTU	Purple	1	600
Flow Cell Flush	FCF	Clear cap, light blue label	1	15,500

Note: We are in the process of reformatting our kits with single-use tubes into a bottle format.

Single-use tubes format:

Bottle format:

Note: This Product Contains AMPure XP Reagent Manufactured by Beckman Coulter, Inc. and can be stored at -20°C with the kit without detriment to reagent stability.

Note: The DNA Control Sample (DCS) is a 3.6 kb standard amplicon mapping the 3' end of the Lambda genome.

Contents	Volume (µl)	No. of tubes	No. of uses
Wash Mix (WMX)	15	1	6
Wash Diluent (DIL)	1,300	2	6
Storage Buffer (S)	1,600	2	6

• Wash Mix (WMX) contains DNase I.
• Wash Diluent (DIL) contains the exonuclease buffer that maximises activity of the DNase I.
• The Storage Buffer allows flow cells to be stored for extended periods of time.

Name	Acronym	Cap colour	No. of vials	Fill volume per vial (μl)
Flow Cell Flush	FCF	6	Clear cap, light blue lable	8,000
Flow Cell Tether	FCT	1	Purple	200

Name	Acronym	Cap colour	No. of vials	Fill volume per vial (μl)
Elution Buffer	EB	Black	2	500
Sequencing Buffer	SB	Red	2	700
Library Solution	LIS	White cap, pink label	2	600
Library Beads	LIB	Pink	2	600
Flow Cell Flush	FCF	Light blue label	2	8,000
Flow Cell Tether	FCT	Purple	2	200

Approximately 6 million isolated white blood cells must be prepared from 1.6 ml of whole blood to use as input in the ultra-long DNA experiment.

Users may isolate white blood cells by any means they think are most appropriate for the whole blood sample to be used. If an alternative method is used, this step can be skipped and proceed directly to the next section of the protocol.

As explained in the introduction, the ultra-long DNA experiment must be performed in duplicate and we recommend performing this step for both volumes of blood side by side. Note: ensure both blood volumes are from the same sample.

In this step, ultra-long DNA is extracted from the isolated cells and quantified before going into the library preparation step.

As explained in the introduction, the ultra-long DNA experiment must be performed in duplicate and we recommend performing this step for both white blood cells isolated from whole blood side by side.

In this step, the extracted ultra-long DNA is prepared for sequencing by tagmentation and rapid attachment of sequencing adapters.

As explained in the introduction, the ultra-long DNA experiment must be performed in duplicate. This step can be performed side by side or separately for both duplicates of 750 µl of extracted uHMW gDNA in EEB.

Once the ultra-long DNA library has been prepared, the PromethION Flow Cell can be primed, and the library prepared with the final sequencing reagents before loading for sequencing to begin. Due to the viscosity of the library, the flow cell priming and loading steps have been modified.

As explained in the introduction, the ultra-long DNA experiment must be performed in duplicate. This step can be performed side by side or separately for both duplicates.

We recommend reloading your PromethION Flow Cell with a fresh ultra-long DNA library to maintain high output, using the modified method for reloading a viscous library.

For the Ultra-long DNA experiment, up to five libraries prepared using the Ultra-Long DNA Sequencing Kit V14 (SQK-ULK114) can be loaded on the PromethION Flow Cell during a sequencing run. We recommend washing the flow cell when ~20-25% of active pores are remaining, which typically occurs after ~20-24 hours of sequencing. Washing removes most of the initial library as well as unblocking pores to prepare the flow cell for loading a new library for further sequencing.

Navigate to the pore activity or the pore scan results plot to see pore availability. Below is an example of pore states observed on a flow cell before and after wash steps are performed. The red asterisks indicates the reloads.

Due to the viscosity of the library, the flow cell washing and reloading steps have been modified. It is also recommended to remove the waste fluid before washing the flow cell and before reloading of an ultra-long DNA library after each priming step.

As explained in the introduction, the ultra-long DNA experiment must be performed in duplicate. This step can be performed side by side or separately for both duplicates.

Once you have loaded your flow cell, the sequencing run can be started on MinKNOW, our sequencing software that controls the device, data acquisition and real-time basecalling. For more detailed information on setting up and using MinKNOW, please see the MinKNOW protocol.

We recommend first basecalling in real-time using the fast basecaller on MinKNOW using the PromethION 24 or 48 device. MinKNOW can be used and set up to sequence in multiple ways:

• On a computer either directly or remotely connected to a sequencing device.
• Directly on a PromethION 24/48 sequencing device.

After real-time basecalling, rebasecall the data using the super-accurate (SUP) basecaller v5.0 model (or newer) in Dorado.

For more information on using MinKNOW on a sequencing device, please see the PromethION 24/48 user manual.

---

Real-time sequencing

To start a run on MinKNOW to sequence ultra-long DNA

1. Navigate to the start page and click Start sequencing.

2. Fill in your experiment details, such as name and PromethION Flow Cell position and sample ID.

3. Select the Ultra-Long DNA Sequencing Kit V14 (SQK-ULK114) on the Kit page.

4. Configure the sequencing parameters as follows: Basecalling: on Modified bases: off Model: fast basecalling Barcoding: off Alignment: off Adaptive sampling: off Advanced options: default settings

5. Configure the data targets as follows: Run duration: 140 hours

6. Configure the analysis workflow: Workflow: off

7. Configure the output parameters as follows: Basecalled output type: .BAM Based on: Time elapsed Frequency: Every 10 minutes FASTQ options - Compression: on Raw reads: on POD5: on FAST5: off

8. Configure the filterning options as follows: Filtering: on Min Qscore: 10 Min read length (kb): 1

Once sequencing is complete, rebasecall your data using the super-accurate (SUP) basecaller using command line Dorado with the following commands as described on the Dorado Github page:

$ dorado basecaller --min-qscore 10 sup ultralong_pod5s/ > ulk_reads.bam

Note: when running Dorado, we recommend stopping other basecalling for the best performance by maximising available memory to Dorado. This can be stopped and restarted when Dorado has finished via the GUI on MinKNOW.

Please remember to change the output file name to differentiate between each of the ultra-long experiment flow cells run in duplicate.

In the Downstream analysis section, we outline further options for analysing your basecalled data for the telomere-to-telomore experiment.

An input of 5 μg of gDNA must be prepared for the Assembly Polishing Kit experiment. Below we outline how to use the QIAGEN Puregene Cell Kit to extract your input.

An input of 5 µg of gDNA must be prepared for the Assembly Polishing Kit experiment. Below we outline how to use the QIAGEN Puregene Cell Kit to extract your input.

After DNA extraction from either cell culture or whole blood, we recommend shearing your DNA to 10 kb before library preparation.

The library is prepared for sequencing using the Assembly Polishing Kit, where DNA nicks are repaired, and a single cycle polymerase step is performed to help achieve haplotype resolved Q50 assemblies. Finally, rapid adapters are attached to the DNA for sequencing.

Once the library has been prepared, the PromethION Flow Cell can be primed before the library is combined with the sequencing reagents and loaded into the flow cell.

Once you have loaded your flow cell, the sequencing run can be started on MinKNOW, our sequencing software that controls the device, data acquisition and real-time basecalling. For more detailed information on setting up and using MinKNOW, please see the MinKNOW protocol.

We recommend first basecalling in real-time using the fast basecaller on MinKNOW using the PromethION 24 or 48 device. MinKNOW can be used and set up to sequence in multiple ways:

• On a computer either directly or remotely connected to a sequencing device.
• Directly on a PromethION 24/48 sequencing device.

After real-time basecalling, rebasecall the data using the assembly polishing model in Dorado.

For more information on using MinKNOW on a sequencing device, please see the PromethION 24/48 user manual.

---

Real-time sequencing

To start a run on MinKNOW to sequence the Assembly Polishing Kit library:

1. Navigate to the start page and click Start sequencing.

2. Fill in your experiment details, such as name and PromethION Flow Cell position and sample ID.

3. Select the Assembly Polishing Kit V14 (SQK-APK114) on the Kit page

4. Configure the sequencing parameters as follows: Basecalling: off Modified bases: off Model: N/A Barcoding: off Alignment: off Adaptive sampling: off Advanced options: default settings

5. Configure the data targets as follows: Run targets: Run limit Action: Stop run when Condition: Flow cell is Value: End of life

6. Configure the analysis workflow: Workflow: off

7. Configure the output parameters as follows: Basecalled output type: .BAM Based on: Time elapsed Frequency: Every 10 minutes FASTQ options - Compression: on Raw reads: on POD5: on FAST5: off

8. Configure the filterning options as follows: Filtering: on Min Qscore: 10 Min read length (kb): 1

1. Ensure you have installed Dorado 0.7.2 from the installation section of the Dorado Github page.

2. Download the the custom assembly polishing command and the basecalling model, and rebasecall your data as described on the Dorado Github page by using the commands outlined below:

$ dorado download dna_r10.4.1_e8.2_apk_sup@v5.0.0

$ dorado basecaller --min-qscore 10 dna_r10.4.1_e8.2_apk_sup@v5.0.0 pod5s/ > apk_reads.bam

Note: when running Dorado, we recommend stopping other basecalling for the best performance by maximising available memory to Dorado. This can be stopped and restarted when Dorado has finished via the GUI on MinKNOW.

In the Downstream analysis section, we outline further options for analysing your basecalled data for the telomere-to-telomore experiment.

Before starting the Pore-C experiment, a custom SPRI bead suspension needs to be prepared. This will be used to deplete non-chimeric monomers and to maximise the frequency of chimeric Pore-C polymers, improving purity ratios and read lengths before library preparation.

Before starting the Pore-C DNA extraction, the white blood cells must be isolated from whole blood whilst maintaining cell viability. Approximately 5–10 ml of whole blood should yield sufficient white blood cells for the Pore-C DNA extraction. If necessary, combine multiple aliquots of whole blood to achieve a final 5–10 ml pooled sample. Using the below method, approximately 10 million white blood cells are prepared in aliquots of 1x white blood cells supplemented with 2% FBS. Approximately 10 million PBMCs are taken forwards into the Pore-C sample preparation step.

Users may isolate white blood cells by any means they feel are most appropriate for the whole blood sample to be used, provided that:

• white blood cells are isolated as soon as possible from fresh blood and no later than 24 hours.
• white blood cells are isolated using a method optimised for cell viability.
• The whole blood is not mixed with any additives, except for anticoagulants (e.g. K2-EDTA), which are acceptable and will not interfere with the Pore-C DNA extraction.

Pore-C extraction is performed across three days to stabilise the three-dimensional interactions of the DNA in the nucleus before extracting the DNA.

During day 1, the PBMCs are prepared for stabilising of the three-dimensional interactions of DNA in the nucleus by chemically cross- linking DNA and protein. The nuclei are then permeabilised to expose the crosslinked cytoskeleton cage and nuclear structures before the chromatin is denatured. The DNA is now accessible to the chosen restriction enzyme† which passively diffuses through the crosslinked cytoskeleton cage and nuclear structures to digest the genome at compatible recognition sites. The sample is incubated overnight which creates clusters of DNA fragments held in proximity by crosslinks between DNA and the cytoskeleton, preserving the original interactions which were crosslinked.

†This protocol has been written using NlaIII and the heat denaturation method as our investigations have found this 4-cutter is particularly suitable for Pore-C across many different species, yielding Pore-C extracts with high contact densities. For more information, please see the "Protocol considerations" section of our Restriction enzyme Pore-C info sheet.

During day 2, the restriction enzymes are heat inactivated to prevent re-digesting ligated products. DNA ligase is added to the clusters of crosslinked DNA and passively diffuses through the crosslinked cytoskeleton cage to ligate the cohesive ends of proximal monomers into chimeric Pore-C polymers. After ligation, the ligated products can be released from the crosslinked cytoskeleton cages by an overnight proteinase K digestion. This releases the chimeric Pore-C polymers into solution as dsDNA.

During day 3, the chimeric Pore-C dsDNA polymers are purified from the solution of polypeptide fragments and residual reaction buffers. The peptides are removed by using a phenol:chloroform extraction, followed by an ethanol precipitation to purify the DNA from the residual reaction buffers and phenol. The final Pore-C DNA extract is a pool of chimeric dsDNA polymers made of multiple ligated monomers which are sequenced to determine DNA interactions, proximity in sequence space and the three-dimensional structures of chromatin within the nucleus.

In this step, the extracted Pore-C DNA is prepared for sequencing by repairing any nicks in the DNA and preparing the ends for sequencing adapter attachment.

Once the Pore-C DNA extracts have been prepared using the Ligation Sequencing Kit V14 (SQK-LSK114), the PromethION Flow Cell can be primed, and the library prepared with the final sequencing reagents for the first library load to be sequenced.

Once you have loaded your flow cell, the sequencing run can be started on MinKNOW, our sequencing software that controls the device, data acquisition and real-time basecalling. For more detailed information on setting up and using MinKNOW, please see the MinKNOW protocol.

We recommend basecalling in real-time using the high-accuracy (HAC) basecaller on MinKNOW using the PromethION 24 or 48 device. MinKNOW can be used and set up to sequence in multiple ways:

• On a computer either directly or remotely connected to a sequencing device.
• Directly on a PromethION 24/48 sequencing device.

For more information on using MinKNOW on a sequencing device, please see the PromethION 24/48 user manual.

---

Real-time sequencing

To start a run on MinKNOW to sequence Pore-C DNA

1. Navigate to the start page and click Start sequencing.

2. Fill in your experiment details, such as name and PromethION Flow Cell position and sample ID.

3. Select the Ligation Sequencing V14 (SQK-LSK114) on the Kit page.

4. Configure the sequencing parameters as follows: Basecalling: on Modified bases: off Model: High-accuracy (HAC) basecalling Barcoding: off Alignment: off Adaptive sampling: off Advanced options: default settings

5. Configure the data targets as follows: Run duration: 72 hours

6. Configure the analysis workflow: Workflow: off

7. Configure the output parameters as follows: Basecalled output type: .BAM & .FASTQ Based on: Time elapsed Frequency: Every 10 minutes FASTQ options - Compression: on Raw reads: on POD5: on FAST5: off

8. Configure the filterning options as follows: Filtering: on Min Qscore: 9 Min read length (kb): 0.2

Your Pore-C experiment data is basecalled live in MinKNOW during sequencing, using the high-accuracy (HAC) basecaller.

In the Downstream analysis section, we outline further options for analysing your basecalled data for the telomere-to-telomore experiment.

Assembly workflow

To carry out the telomere-to-telomere downstream analysis, we recommend previous bioinformatics experience. The commands below combine all three datasets from the ultra-long DNA, Pore-C, and Assembly Polishing Kit experiments.

More information about the workflow and examples of the inputs and outputs of this analysis protocol can be found on the nanopore-only telomere-to-telomere dataset and blog post on the EPI2ME page.

An EPI2ME workflow combining all three datasets will be released in the future to give more accessibility for users of all experience. More information about this release will be published on the Community.

Note: all basecalling must have been completed before starting downstream analysis.

Setting up the environment

The Nanopore-only telomere-to-telomere assembly workflow is summarised in the figure above and requires the following tools and their dependencies to be installed:

• Dorado (≥v0.7.3) for read correction
• Verkko (= v2.1) for assembly
• Medaka (≥v1.12.1) for polishing
• Minimap2 (≥v2.27) for alignment
• Samtools (≥v1.16) for format conversion

See the appendix below for an example how to set up a suitable environment. For Dorado, please follow installation instructions on the Github page.

For more information about the workflow and for examples of the inputs and outputs of this analysis protocol, please see the Nanopore-only T2T dataset. The basecalled data from s3://ont-open-data/londoncalling2024/assembly/basecalling/ may be downloaded to test the full workflow. Instructions to download the dataset from S3 are outlined in the EPI2ME blog post.

Read filtering

Basecalled reads for your Ultra-long DNA experiment (unaligned .bam) should be filtered for reads with mean quality of ≥ 10. We also recommend filtering for ≥ 10kb read length to speed up run time but this is optional.

Dorado reports read mean quality as a “qs” tag in the output .bam file. The command to filter ULK reads by quality and read length is:

samtools view -@ <threads> -b \ 
 -e '[qs] >= 10 && length(seq) > 10000' \ 
 input.bam > output.bam  

Note: If not already done, the command to filter APK and Pore-C reads by quality is:

samtools view -@ <threads> -b -e '[qs] >= 10' input.bam > output.bam 

Ultra long read correction

First, pass the basecalled Ultra Long (SQK-ULK114) reads through read correction in Dorado.

To do this, the basecalled Ultra Long BAMs should first be combined into a single BAM:

samtools merge -@ <threads> -o ulk_reads.bam \ 
  ulk_reads_flowcell_1.bam ulk_reads_flowcell_2.bam 

This merged ULK BAM should then be converted to fastq:

samtools fastq -@ <threads> \ 
 ulk_reads.bam > uncorrected_ulk_reads.fastq 

Then passed through Dorado read correction:

dorado correct uncorrected_ulk_reads.fastq > corrected_ulk_reads.fasta 

Assembly instructions

1. Assemble the corrected reads using the Verkko assembler.

1.1. The basecalled Pore-C BAM should also be converted to fastq for this step:

samtools fastq -@ <threads> porec_reads.bam > porec_reads.fastq 

1.2. Run Verkko using the following command :

verkko –d asm \ 
--hifi corrected_ulk_reads.fasta \ 
--nano uncorrected_ulk_reads.fastq \  
--porec porec_reads.fastq \
--no-correction

2. Having run Verkko in its entirety, the last step is to polish the Verkko assembly (assembly.fasta) using data from the APK sequencing kit.
This is done using the medaka_consensus_joint script which simultaneously uses both ULK and APK data for error correction and will be in the path in the Medaka environment.

medaka_consensus_joint \ 
   -i "${APK_BAM}" -v apk -i "${ULK_BAM}" -v ulk \ 
   -t ${THREADS} -o "${OUTPUT}" \ 
   -m r1041_e82_260bps_joint_apk_ulk_v5.0.0 \ 
   -d "${VERKKO_ASSEMBLY}" 

Appendix

Example: Installing required software with conda

0. Install conda (or mamba, micromamba) if you do not already have it.

1. Add the following to a file called t2t.yml
name: t2t
channels: - conda-forge - bioconda - nvidia
dependencies: - python=3.9 - verkko=2.1 - minimap2>=2.27 - samtools>=1.16 - libcublas - bcftools>=1.16 - pip - pip: - medaka >= 1.12.1

2. To create the environment:
conda env create --file=t2t.yml

3. To activate the environment:
conda activate t2t

We also have an FAQ section available on the Nanopore Community Support section.

If you have tried our suggested solutions and the issue still persists, please contact Technical Support via email (support@nanoporetech.com) or via LiveChat in the Nanopore Community.

We also have an FAQ section available on the Nanopore Community Support section.

If you have tried our suggested solutions and the issue still persists, please contact Technical Support via email (support@nanoporetech.com) or via LiveChat in the Nanopore Community.