Mar 10, 2026

Bravo workstation: automated ancient DNA extraction part II: DNA purification (binding buffer D) V.1

DOI
https://dx.doi.org/10.17504/protocols.io.eq2ly4zwrlx9/v1
  • 1Max Planck Institute for Evolutionary Anthropology
  • MPI EVA Ancient DNA Core Unit
Ancient DNA Core Unit
Icon indicating open access to content
QR code linking to this content
DOI: https://dx.doi.org/10.17504/protocols.io.eq2ly4zwrlx9/v1
Protocol Citation: Ayinuer Aximu Petri, Anna Schmidt, Sarah Nagel, Elena Essel, Matthias Meyer 2026. Bravo workstation: automated ancient DNA extraction part II: DNA purification (binding buffer D). protocols.io https://dx.doi.org/10.17504/protocols.io.eq2ly4zwrlx9/v1
License: This is an open access  protocol  distributed under the terms of the  Creative Commons Attribution License,  which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Protocol status: Working
We use this protocol and it's working
Created: July 01, 2025
Last Modified: March 10, 2026
Protocol  Integer ID: 221371
Keywords: second step of ancient dna extraction, ancient dna extraction, ancient dna extraction part ii, ancient dna extraction part, automated ancient dna extraction part ii, dna purification, dna library preparation, workflows of the ancient dna core unit, ancient dna core unit, stranded dna library preparation, dna extract, ultrashort dna fragment, preparation of lysate, purification, lysate preparation, extraction method, volumes for downstream library preparation, downstream library preparation, cap tube, sediment, stranded dna
Funders Acknowledgements:
Max Planck Society
Grant ID: -
Abstract
Following the preparation of lysates (see Nagel et al. 2025), this protocol describes the second step of ancient DNA extraction - DNA purification - from bones, teeth and sediments, building on the silica-based extraction method described by Dabney et al. 2013 and refined by Rohland et al. 2018. Purification is performed using binding buffer 'D' (Rohland et al. 2018). The protocol includes specific adaptations to the clean room environment and workflows of the Ancient DNA Core Unit at the MPI-EVA. Lysate aliquots of 150 µl each should be provided in screw-cap tubes, as detailed in Nagel et al. 2025. After purification, DNA extracts will be delivered in screw-cap tubes in 30 µl volumes for downstream library preparation (e.g., see Nagel et al. 2026).

References
Dabney, J., Knapp, M., Glocke, I., Gansauge, M.-T., Weihmann, A., Nickel, B., Valdiosera, C., García, N., Pääbo, S., Arsuaga, J.-L., & Meyer, M. (2013). Complete mitochondrial genome sequence of a Middle Pleistocene cave bear reconstructed from ultrashort DNA fragments. Proceedings of the National Academy of Sciences of the United States of America, 110(39), 15758-15763.

Nagel, S., Schmidt, A., Meyer, M. & Essel, E. (2025). Ancient DNA extraction part I: lysate preparation. protocols.io https://dx.doi.org/10.17504/protocols.io.4r3l212oqg1y/v1

Nagel, S., Schmidt, A., Aximu-Petri, A., Patova, A., Szymanski, M., Essel, E. & Meyer, M. (2026). Bravo workstation: automated single-stranded DNA library preparation (ssDNA2.0). protocols.io https://dx.doi.org/10.17504/protocols.io.kqdg32bdpv25/v2

Rohland N, Hofreiter M. Ancient DNA extraction from bones and teeth. Nat Protoc. 2007;2(7):1756-62. doi: 10.1038/nprot.2007.247. PMID: 17641642.
Materials
Materials

ReagentSupplierCat. no.Decont. *
Large-volume buffers/reagents
Binding Buffer D 5M40 †self - UV
TET †self - UV
PE **Qiagen19065UV
Other reagents
3M sodium acetate pH 5.2Sigma Aldrich S7899-500ML UV
Silica magnetic beads, 5 mlG-Biosciences 786-915 -
Extraction positive controls †self - -
Purification negative controls †self - -
Consumables
Bravo 96LT 250 µl sterile, filtered tipsAgilent 19477-022 -
FluidX tubes 0.8 ml external thread jacketBrooks68-0801-11 -
FluidX cap carrier with caps, 96 wellBrooks68-53111-50N -
1.5 ml LoBind tubesEppendorfVB-0285UV
Filter tip, Natur, 1250 μl, low retentionGreiner Bio-One778363 -
Filter tip, Natur, 300 μl, low retentionGreiner Bio-One775353 -
5 ml glass pipetteGreiner Bio-One357543 -
96 Matrix open reservoir movableThermo Scientific1064156 -
HTS deep well reservoirsKisker97813 -
96 Deep Well PP Sqr Well PlateAxygenP-2ML-SQ-C-S -
Twin.tec PCR plates 96 LoBind, skirtedEppendorfVB-0433 -
Microseal B Adhesive foilBio-RadMSB1001 -
* See documents in the Appendix for decontamination instructions.
† See documents in the Appendix for preparation of buffers and controls.
** Fill up concentrate as indicated in the description on the bottle before decontamination.

Equipment

  • Bravo-B NGS workstation G5522A with 96-channel LT pipette head and associated equipment (red thermal plate insert for 96-well PCR plates, 96-well magnetic rack) for DNA extraction
  • Hood for NGS workstation B with HEPA-filtered ventilation system and strip curtain (custom-built, optional)
  • FluidX rack barcode reader (Brooks Life Sciences, cat. no 20-4018)
  • Tube decapper (Aperio 8-Channel Semi- Automatic Screw Top tube rack decapper, Brooks Life Sciences, cat. no. 46-6502)
  • Centrifuge for PCR plates (e.g., Eppendorf, cat. no. 5948000913)
  • Table-top micro-centrifuge for 1.5 ml tubes (e.g., Carl Roth Mini-Zentrifuge ROTILABO, cat. no. T464.1)
  • Magnetic rack for Eppendorf tubes (e.g., MagJET Separation Rack, ThermoFisher Scientific, cat.no. MR02)
  • Serological pipette controller (e.g. battery-powered pipetting aid ROTILABO, cat. no. TC16.1)

Bravo electronic protocol files

Available on Github: https://github.com/adnacore/Bravo-workstation-DNA-extraction


Appendix
Documents for extraction controls

Document
NAME

Extraction positive controls (EPCs)

CREATED BY
Ancient DNA Core Unit

Document
NAME

Purification negative controls (PNCs) prepared on the Bravo NGS Workstation

CREATED BY
Ancient DNA Core Unit
Documents for buffers
Document
Binding buffer D (5M40)
NAME

Binding buffer D (5M40)

CREATED BY
Ancient DNA Core Unit

Document
NAME
TET buffer
CREATED BY
Sarah Nagel
Documents for decontamination procedures
Document
NAME
UV decontamination of reagents/buffers
CREATED BY
Elena Essel

Document
NAME
UV decontamination of materials
CREATED BY
Elena Essel

Safety warnings
For information about potential health and safety hazards, please refer to the Material Data Safety Sheets associated with the reagents and chemicals used.
Reagent and sample preparation (days/weeks/months before the run)
Prepare reagents (sodium acetate), buffers (Binding buffer D, TET, PE) and aliquots of extraction positive controls (EPCs) and (optionally) purification negative controls (PNCs) as described in MATERIALS.
Decontaminate reagents and consumables as described in MATERIALS.
Design a new DNA extraction experiment by generating an extraction summary table and name it ExPl_XXX in chronological order. Follow the guidelines below.

Note
[Note]
Setup the extraction summary table in MS Excel for a FluidX sample rack containing 96 FluidX screw-cap tubes with sample DNA lysates and their corresponding extraction negative controls (ENCs). Reserve positions for extraction positive, library negative and library positive controls (EPCs, LNCs and LPCs). Observe the following rules:
  • Use the sample lysate information from the tab "Lysates" of the "samples_to_screen" list available at "P:\AncientDNA\samples" to design the plate setup
  • By default, reserve four positions each for controls so that each quarter of the sample rack (i.e., columns 1-3, 4-6, 7-9 and 10-12) carries a set of controls (EPCs, LNCs and LPCs)
  • LNCs and LPCs will be added later during library preparation. Use empty FluidX tubes as spaceholders at those positions when running the extraction
  • In case there are no corresponding ENCs in the extraction setup (external samples), use purification negative controls (PNCs)
  • Re-position controls between different extraction runs for unbiased monitoring of DNA extraction efficiency and contamination across the extraction plate
  • Save the summary sheet on the coreunit server: \SummaryTables\Extractions\year
  • A template file is provided below.
Download TEMPLATE_ExPl_XX_SummaryTable_JJJJMMDD.xlsxTEMPLATE_ExPl_XX_SummaryTable_JJJJMMDD.xlsx276.1KB

Before each plate assembly, create an entry in Labfolder and document the experiment.

Note
[Note]
The Labfolder entry name consists of the name of the Labfolder template and the extraction plate ID. To document the plate assembly fill the provided data element fields in Labfolder.

Prepare a FluidX sample rack containing 96 sample DNA lysates and extraction negative and positive controls in FluidX screw-cap tubes in 150 µl volumes. Prepare the sample DNA lysate rack as follows:
Start the FluidX barcode reader software and allow the camera to initialize.
Add FluidX tubes containing lysates and controls, as well as empty tubes as spaceholders for LNCs and LPCs in library preparation, to the sample rack. Use the information from the respective extraction summary table available on the coreunit server: \SummaryTables\Extractions\year.

Note
[Documentation]
Note down the batch information for the EPCs in the respective field in Labfolder.

Use the FluidX barcode reader to read the bottom barcodes of all tubes in the rack and the rack ID in the front.

Note
[Documentation]
Make sure the positions of the FluidX tubes match the positions defined in the extraction summary table by comparing the tube IDs.

[Labelling]
Label the plate with "Lysates for ExPl_XXX", date and initials.

Store the rack at:
  • room temperature for same day DNA extraction
  • 4 °C for DNA extraction within 4 weeks
  • -20 °C for long term storage
Preparing and starting a DNA extraction
Before each run, create an entry in Labfolder and document the experiment.

Note
[Note]
The Labfolder entry name consists of the name of the Labfolder template and the extraction plate ID. To document your experiment fill the provided data element fields in Labfolder. They later serve as a template for creating the CoreDB entry.

Take the FluidX sample rack from its storage location. If the rack has been stored in the freezer, set a ThermoShaker to 37 °C and 200 rpm and place the FluidX sample rack to the ThermoShaker for a minimum of 10 min.

Note
[Note]
In case the FluidX sample rack has been stored in the freezer, visually verify that all lysates are completely thawed by randomly selecting individual tubes from the middle of the rack. Let the rack incubate until thawing is complete.

To a 250 ml bottle containing 157.5 ml of binding buffer, add 6.3 ml of sodium acetate. Mix by shaking the bottle.

Note
[Note]
Binding buffer and sodium acetate may cause corrosive damage to laboratory equipment. Therefore, only use the pipette dedicated for pipetting sodium acetate and binding buffer.

Prepare silica beads:
Fully resuspend stock suspension of silica magnetic beads by vortexing thoroughly. Make sure no bead pellet remains.

Note
[Documentation]
Note down the batch information about the beads in the respective fields in Labfolder.

Transfer 1,050 μl stock suspension of beads into an Eppendorf 1.5 ml Safe-lock tube. Pellet beads in magnetic rack, pipette off and discard supernatant.

Note
[Note]
Pelleting the beads in the magnetic rack will take several minutes. Make sure that the pellet has formed properly and that the supernatant is almost clear.

Wash beads twice: Add 1,000 µl TET buffer, vortex, place the tube on the magnetic rack, and pipette off and discard the supernatant.

Note
[Note]
Silica beads tend to pellet strongly in the tube. Make sure to properly mix the beads when washing. Do not centrifuge the tube after mixing to avoid the formation of pellets at the bottom of the tube.

[Documentation]
Note down the buffer batches in the respective fields in Labfolder.

Resuspend the beads in 1,000 μl of the binding buffer/sodium acetate mixture from by vigorous vortexing. Briefly centrifuge the tube. 

Transfer the bead suspension into the binding buffer mixture.

Note
[Note]
Make sure to transfer the entire bead suspension to the binding buffer/sodium acetate mixture. If necessary, rinse the 1.5 ml tube with binding buffer/sodium acetate mixture until all beads are carried over.

Switch on all components of the Bravo system, including the external cooling device (set to 4 °C). Switch on light and ventilation ("Betrieb") inside the robot hood.
Log into the VWworks software using the administrator account (password "a"). Load the DNA extraction preparation protocol under "S:\Bravo_protocols\MPI-EVAN-homebrew\forms\DNA_extraction\V2_Extraction_purification_96.VWForm". Initialize the system.
Set up Bravo deck as indicated by the form file. The MiniHub can be filled later, when the robot is already running.
Unpack 16 Bravo tip boxes. Remove the lids and load the tip boxes into stackers 1-2 of the BenchCel as indicated by the form file.

Note
[Note]
To save time, it is possible to load only one tip box into each stacker at this point and add the remaining ones directly after the run started.

Select the proper settings in the form file:
  • Protocol: 5M40_DNA_Extraction_96
  • Elution volume: 15 µl (two elution steps with 15 µl for a final extract volume of 30 µl)
  • Plate type for lysate: 0.8 ml FluidX tubes
  • Plate type for final extract: 0.8 ml FluidX tubes
  • Reservoir type for Binding buffer: HTS
  • Reservoir type for PE and TE buffer: HTS
Vortex the sample rack from and briefly spin down the liquids in the plate centrifuge at 3,000 - 4,000 rpm. Open the FluidX screw-cap tubes in the sample rack with the decapper and discard the lids. Place the sample rack into the Bravo deck as indicated by the form file.
Start run by clicking the "Run" button and follow the instructions by VWorks. Load remaining reagents and consumables ( - ) within 40 min after starting the run.

Note
[Note]
Run time is approximately 3 h. The first 15 min are used for aliquoting the bead suspension. Afterwards, the robot will transfer the lysate to the bead suspension. Once the tips are filled with lysate, the robot will pause for ~ 3 min to allow for visual inspection of pipetting success. Thoroughly monitor this part of the run. Make sure that the robot pipettes correctly and that no tubes remain stuck on the robot head during lysate transfer. If tubes become stuck, gently shake the tube rack by touching it from the right side of the Bravo deck to detach the tubes from the head.
CAUTION: Do not touch the tube rack from the front side of the Bravo deck. Doing so might interfere with the ongoing protocol and could cause lysate to leak from the tips.

Fill buffer reservoirs as listed in the table below and load them into the MiniHub as indicated by the form file.

BufferVolume (ml)
PEapprox. 60
TETapprox. 30

Note
[Documentation]
Note down the buffer batches in the respective fields in Labfolder.

Label the wash plate and the elution plate (two empty, full-skirted 96-well LoBind PCR plates) with extraction plate ID, 'wash plate'/'elution plate', date and your initials. Place them into the MiniHub as indicated by the form file.
Take an empty FluidX rack for the final extracts. Use the FluidX barcode reader to read the bottom barcodes of all tubes in the rack and the rack ID in the front. Save the tube and rack barcodes in your extraction summary table.
Label the empty FluidX rack with extraction plate ID, date and your initials, remove the lid and place the rack into the decapper to open the FluidX tubes. Store the lids on a clean 96-well lid holder, which can be found next to the decapper. The lids will be used later to close the FluidX sample rack containing the final DNA extracts. Place the uncapped FluidX rack into the the MiniHub as indicated by the form file.
If applicable, add the missing tip boxes to the stackers. Double-check that all required positions are correctly filled.
Finishing the DNA extraction
When the run is finished, seal the elution plate with adhesive foil while keeping it on the magnet. Remove the plate and inspect it for any bead carryover. If the plate contains beads, store it in the freezer for 6 months in case it is needed for trouble-shooting. If not, discard the elution plate.

Note
[Note]
To avoid cross-contamination among plate positions caused by possible electrostatic movements of the beads, do not remove the elution plate from the magnet before sealing.

Seal the wash plate with adhesive foil and store the plate for 6 months in the freezer. 
Carefully close the FluidX tubes containing the final eluate using the lids stored on the 96-well lid holder and the decapper. Store the Fluidx rack in the freezer until library preparation.
Discard all the plastic ware that is still in the robot, except the liquid waste in the square well plates. When doing so, check the lysate FluidX rack, plates and reservoirs for any remaining liquids to make sure that all pipetting steps were performed accurately.

Note
[Documentation]
In case of unexpected remaining volumes in the disposed plastic ware, add comments in the 'notes' section of Labfolder as well as in the documentation of the experiment in Mattermost.

Pour the liquid waste from the square well plates into 500 ml labelled bottles next to the sink using a funnel. Discard used tips to regular trash and the tip boxes to the large waste bin on the left side of the cooling unit.
Clean the robot by wiping the Bravo deck and the benches using tissues and ethanol. Switch off the robot and the cooling device and restart the computer so that remote access to the computer becomes possible. Switch off the computer screen.
After successful extraction, finalize the extraction summary sheet and update the "samples-to-screen" list. Follow the guidelines below.

Note
[Documentation]

  • Create extract IDs in CoreDB. If samples are registered in Pandora, create Pandora extract IDs as well.
  • Add the extract IDs to the extraction summary table and copy all rows from the extraction summary table into the tab "Extracts" of the "samples-to-screen" list stored at "P:\AncientDNA\samples"
  • Delete the rows containing the information of all lysates converted into extracts from the tab "Lysates" of the "samples-to-screen" list, save a copy of the updated list with date and initials and put the old versions in the "screening_list_backup" folder.