Identification of Accessible Chromatin Regions with MNase-seq

Materials and reagents

Materials

1. Wheat seedlings at the three-leaf stage (Aikang58, CAAS)

2. Miracloth (Calbiochem, catalog number: 475855-1R)

3. 50 mL centrifuge tube (Corning, catalog number: 430290)

4. 5 mL pipette (Corning, catalog number: 4487)

   
   

1. PIPES (BBI, catalog number: A600719)

2. Sorbitol (BBI, catalog number: A610491)

3. EGTA (Millipore, catalog number: 324626)

4. DTT (Thermo Fisher, catalog number: R0861)

5. Spermine (Sigma-Aldrich, catalog number: S3256)

6. Spermidine (Sigma-Aldrich, catalog number: S2626)

7. 37% Formaldehyde (Sigma-Aldrich, catalog number: 818708)

8. PMSF (Thermo Fisher, catalog number: 36978)

9. Glycine (Diamond, catalog number: A100167)

10. 1 M Tris-HCl (pH 7.5) (Sangon, catalog number: B548124)

11. 0.5 M EDTA (pH 8) (Sangon, catalog number: B540625)

12. Sucrose (Diamond, catalog number: A100335)

13. MgCl₂ (Sigma-Aldrich, catalog number: M8266)

14. CaCl₂ (Sigma-Aldrich, catalog number: C3306)

15. KCl (Sigma-Aldrich, catalog number: P9541)

16. NaCl (Sigma-Aldrich, catalog number: S3014)

17. SDS (Thermo Fisher, catalog number: 28364)

18. Phenol:Chloroform:Isoamyl alcohol 25:24:1 (Wako, catalog number: 311-90151)

19. RNase (Thermo Fisher, catalog number: EN0531)

20. Isopropanol (Sangon, catalog number: A507048)

21. 1× TE buffer (Sangon, catalog number: B548106)

22. Glycerol (Diamond, catalog number: A100854)

23. Triton X-100 (Sigma-Aldrich, catalog number: T8787)

24. Percoll (GE, catalog number: 17-0891)

25. MNase (Thermo Fisher, catalog number: 88216)

26. Proteinase K (Solarbio, catalog number: 17-0891)

27. NEBNext UltraTM DNA Library Prep kit for Illumina (NEB, catalog number: E7645S)

28. Qiaex II gel extraction kit (Qiagen, catalog number: 20021)

    
    

Solutions

1. Nuclei isolation buffer (see Recipes)

2. Fixation buffer (see Recipes)

3. Percoll cushion solution (see Recipes)

4. MNase digestion buffer (see Recipes)

   
   

Recipes

Note: All concentrations listed are final concentrations.

1. Nuclei isolation buffer

   15 mM PIPES (NaOH at pH 6.8)

   0.32 M sorbitol

   80 mM KCl

   20 mM NaCl

   0.5 mM EGTA

   2 mM EDTA

   1 mM DTT

   0.15 mM spermine

   0.5 mM spermidine

   
   

2. Fixation buffer

   Nuclei isolation buffer, add 1% formaldehyde and 0.1 mM PMSF freshly

   
   

3. Percoll cushion solution

   50% (vol/vol) Percoll in nuclei isolation buffer

   
   

4. MNase digestion buffer

   50 mM Tris-HCl at pH 7.5

   320 mM sucrose

   4 mM MgCl₂

   1 mM CaCl₂

Equipment

1. Mortar and pestle (Avantor, catalog number: HALDL55/1/G)

2. Beaker (30 mL) (PYREX, catalog number: 1000-30)

3. Magnetic stirrer (Thermo Fisher, catalog number: S194615)

4. Hybridization oven (Galanz, catalog number: P70J17L-V1)

5. 4 °C centrifuge (Eppendorf, catalog number: 5804)

6. Nanodrop 2000 (Thermo Fisher, catalog number: 13-400-412)

Software

1. Deeptools v3.4.3 (Max Planck Institute for Immunobiology and Epigenetics; https://deeptools.readthedocs.io/en/latest/index.html) (February 2022)

2. Samtools v1.6 (Genome Research Limited; http://www.htslib.org/doc/samtools.html) (February 2022)

3. BEDTools v2.29.1 (University of Utah; https://bedtools.readthedocs.io/en/latest/content/bedtools-suite.html) (February 2022)

4. R v3.2.2 (Lucent Technologies; https://www.r-project.org/) (March 2022)

5. IGV v2.5.0 (University of California; http://software.broadinstitute.org/software/igv/) (March 2022)

6. Trimmomatic v0.36 (THE USADEL LAB; http://www.usadellab.org/cms/?page=trimmomatic) (February 2022)

7. Bowtie2 v2.3.4 (Johns Hopkins University; http://bowtie-bio.sourceforge.net/bowtie2/index.shtml) (February 2022)

8. FastQC v0.11.8 (Babraham Bioinformatics; https://www.bioinformatics.babraham.ac.uk/projects/fastqc/) (February 2022)

Procedure

1. Material fixation

   1. Sampling: Harvest 10 g of leaves or roots (water rinsing) from the three-leaf-stage wheat seedlings after 21 days of seeds’ germination and immediately freeze the tissues in liquid nitrogen. Store at -80 °C.

   2. Grinding: Ground the frozen materials under liquid nitrogen using a mortar and pestle (previously cooled with liquid nitrogen).

   3. Fixation: Prepare 100 mL of pre-cooled fixation buffer in a beaker: add 2.78 mL of 37% formaldehyde to 1% and 102 μL of 100 mM PMSF to 0.1 mM. Pour the ground sample into the beaker and magnetically stir the sample for 10 min (100–200× g).

   4. Termination: Stop the reaction by adding 6.87 mL of 2 M glycine stock to 125 mM and magnetically stir for another 5 min.

      
      

2. Nuclear extraction

   1. Stirring: Slowly add 12 mL of 10% (vol/vol) Triton X-100 stock to 1%, gently agitate, and incubate for 10 min. Pre-cool the refrigerated centrifuge now.

   2. Filter: Filter the solution into a beaker with two layers of Miracloth filter cloth and add 15 mL of Percoll cushion solution into a 50 mL centrifuge tube. Gently add 35 mL of the filtrate with a pipette on the top of the solution surface and ensure the liquid flows down along the tube wall without disturbing the boundary layer. Centrifuge at 3,000× g for 15 min at 4 °C (acceleration and deceleration should be slow) (see Note 1).

   3. Transfer: The middle layer of the centrifuged Percoll filtrate contains the nuclei; transfer this layer to 50 mL centrifuge tubes (see Note 1) and add a 1-fold volume of MNase digestion buffer. Centrifuge at 2,000× g for 10 min at 4 °C.

   4. Reconstitution: Re-dissolve the precipitate from each 35 mL in 2.5 mL of MNase digestion buffer and dispense the sample into five tubes with 500 μL each. Measure and record the DNA concentration by Nanodrop 2000 (approximately 50–150 ng/µL).

   5. Microscopy: Observe the morphology of the nuclei with a fluorescence microscope. (The nuclei can be flash-frozen in liquid nitrogen and stored at -80 °C for up to one month.)

      
      

3. Digestion and de-crosslinking

   1. Digestion: Thaw proteinase K, MNase, and five tubes of nuclei samples on ice (the nuclei can only be thawed once); then, treat two tube samples with 10 U/mL MNase (light), two tube samples with 100 U/mL MNase (heavy), and one tube sample as a control. Incubate all for 5 min at room temperature (see Note 2).

   2. Termination: Add 10.2 μL of 0.5 M EGTA stock to 10 mM and vibrate the samples to stop the digestion reaction.

   3. De-crosslinking: Add 57.5 μL of 10% SDS to 1% and 5.75 μL of 10 μg/μL proteinase K to 100 μg/mL, vibrate the samples, and incubate overnight in a 65 °C hybridization oven (>6.5 h).

      
      

4. Product extraction

   1. DNA extraction: Add an equal volume of 600 μL of Phenol:Chloroform:Isoamyl alcohol 25:24:1 and mix thoroughly by inversion. Centrifuge at 12,000× g for 10 min at room temperature. Transfer the supernatant into new 1.5 tubes, add 5 μL of RNase to 40 μg/μL, mix by inverting, and incubate at 37 °C for 15 min. Add 400 μL of pre-cooled isopropanol and mix gently. Then, place tubes at -20 °C for 20 min to precipitate DNA. Pre-cool the centrifuge and centrifuge at 12,000× g for 10 min at 4 °C. Discard the supernatant, add 1 mL of 70% ethanol, mix by inverting, and wash twice. Dry the remaining ethanol for 20 min in a fume hood. Dissolve the precipitate in 30 μL of 1× TE buffer.

   2. Fragment selection: Run 1% agarose gel (using single-color loading), select and extract the 100–200 bp DNA fragment (Figure 2, Note 2), and purify the DNA using the Qiaex II gel extraction kit.

      
      

   
   Figure 2. Ladders of MNase-digested DNA bands. The nuclei are treated with different concentrations of MNase and the DNA fraction from 100 to 200 bp is purified from the gel and sequenced.

   
   

5. Library preparation and sequencing

   1. Library construction: Construct the MNase-seq library using the NEBNext^® Ultra^TM DNA Library Prep kit for Illumina^® according to the instructions, where approximately 1 μg of total DNA is used.

   2. High-throughput sequencing: Sequence the libraries on the Illumina platform HiSeq2000 system (or any Illumina sequencing instrument) to obtain 150 bp paired-end reads with 300 million reads (~90 G) for each sample of hexaploid wheat (genome size ~15 G) (see Note 3). Sequence the libraries for two biological replicates.

Data analysis

The analysis of arbitrary MNase-seq data can be achieved through the following instructions, with only minor modifications to the given code. Software used in this protocol can be easily installed via the conda command of anaconda ( https://www.anaconda.com/products/individual) or downloaded from the official website and installed by following the instructions. Make sure that the software is installed and the environment is set up normally in your Linux device before the program is executed.

1. Uploading data: Create a new working directory to store all the MNase-seq-related data and files and save all the raw data to a dedicated folder under that directory.

   
   

   cd data1/user/path # enter your working directory

   mkdir MNase_seq

   cd MNase_seq

   mkdir raw_data # move all your sequencing data to this directory

   
   

2. Quality control: Use FastQC to evaluate the quality of raw data obtained by sequencing, including basic statistics, base and sequence quality, GC content, sequence length distribution, adapter content, and other indicators of data.

   
   

   mkdir ./fastqc_result

   for i in ./raw_data/*.fq.gz

   do

   fastqc -o ./fastqc_result $i

   done

   
   

3. Reads filtering: Use software such as Trimmomatic to filter the raw data to obtain clean data. The main goal of this step is to remove the adapter sequence, leading and trailing low-quality or N bases and sequences too short in length. When renaming the filtered data, tissue, digestion, and replicate should be considered. For example, “leaf_light_rep1” can be used to represent the first replicate with light digestion in leaves, and “root_heavy_rep2” can represent the second replicate with heavy digestion in roots.

   
   

   mkdir clean_data

   java -jar trimmomatic-0.36.jar PE -phred33 \

     ./raw_data/*.r1.fq.gz \

     ./raw_data/*.r2.fq.gz \

     ./clean_data/$name.1P.fq \

     ./clean_data/$name.1U.fq \

     ./clean_data/$name.2P.fq \

     ./clean_data/$name.2U.fq \

     ILLUMINACLIP:data1/user/tools/Trimmomatic-0.36/adapters/\

     TruSeq3-PE-2.fa:2:30:10:1:true \

     LEADING:5 TRAILING:5 MINLEN:20 >trimmomatic.$name.log

   
   

4. Quality control: Use FastQC again to evaluate the quality of the clean data. If there is any undesirability, the parameters of the previous step can be adjusted and re-filtered until the reads can be used for the alignment in the next step.

   
   

   for i in ./clean_data/*.fq.gz

   do

   fastqc -o ./fastqc_result $i

   done

   
   

5. Sequence alignment: The filtered clean data is aligned to the reference genome using Bowtie2 software. Before that, library the genome with bowtie2-build and index the fasta file with Samtools faidx. Use the same parameters as suggested in maize for alignment.

   
   

   bowtie2-build ref_genome.fa ref.genome

   samtools faidx ref_genome.fa

   bowtie2 --phred33 -p 10 --reorder -5 6 \

     --no-mixed --no-discordant --no-unal --dovetail \

     -x /index_path/ref.genome \

     -1 ./$name.1P.fq \

     -2 ./$name.2P.fq \

     -S ./$name.sam >bowtie2.$name.log

   
   

6. Non-unique and duplicate alignments removal: Screen according to the MAPQ value in the alignment results. The larger the MAPQ value, the higher the confidence of the unique alignment. Here, we extract the part of MAPQ > 20. After that, remove the repeated reads due to PCR library amplification. Different software, like Picard, Sambamba, or Samtools, remove PCR duplicates. Here, we used Samtools markdup, before which the paired-end sequencing data were sorted by reads name and coordinates successively. A large index of the bam file should be established after that for the large genome.

   
   

   function samtools_scripts(){

    samtools view -q 20 -bhS $name.sam -o $name.Q20.bam

    samtools sort -n $name.Q20.bam -o $name.Q20.nsort.bam

    samtools fixmate -m $name.Q20.nsort.bam $name.Q20.nsort.ms.bam

    samtools sort $name.Q20.nsort.ms.bam -o $name.Q20.ms.psort.bam

    samtools markdup -r $name.Q20.ms.psort.bam $name.Q20.psort.markdup.bam

    samtools index -c $name.Q20.psort.markdup.bam

   }

   samtools_scripts $name >samtools.$name.log

   
   

7. Differential nuclease sensitivity (DNS) identification: In the code here and later, we take “leaf_light_rep1” and “leaf_heavy_rep1” as examples for demonstration; you can replace the name when running other samples. The mapped reads of heavy digestion are subtracted from light digestion using bamCompare in the deepTools and are then normalized by the CPM method to output a DNS score file in bedgraph format. For this step, a 10-bp bin is used for division and statistics, and then a 50-bp window is used for smoothing.

   
   

   bamCompare -b1 leaf_light_rep1.bam -b2 leaf_heavy_rep1.bam \

     --scaleFactorsMethod None --normalizeUsing CPM --operation subtract \

     --binSize 10 --smoothLength 50 --outFileFormat bedgraph \

     -o leaf_diff_rep1.bdg

   
   

8. Bayes factor calculation: Here, reads coverage from both light-digested and heavy-digested samples is calculated, and the CPM method is used for standardizing; then, the Bayes factor is calculated for reads coverage in 10-bp intervals. You can speed up the program by running splitted files and calculating their Bayes values separately.

   
   

   bamCoverage -b leaf_light_rep1.bam \

     --binSize 10 --smoothLength 50 \

     --normalizeUsing CPM --outFileFormat bedgraph \

     -o leaf_light_rep1.bdg

   
   

   bamCoverage -b leaf_heavy_rep1.bam \

     --binSize 10 --smoothLength 50 \

     --normalizeUsing CPM --outFileFormat bedgraph \

     -o leaf_heavy_rep1.bdg

   
   

   bedtools unionbedg \

     -i leaf_light_rep1.bdg leaf_heavy_rep1.bdg \

     -header -names light heavy \

   >leaf_rep1.unionbdg

   
   

   # run the R script to obtain a file named “leaf_rep1.unionbdg.bayes”

   Rscript bayes_factor_caculator.R leaf_rep1.unionbdg

   
   

9. Accessible chromatin regions identification: Based on the Bayes values in the “leaf_rep1.unionbdg.bayes” file, identify MNase-sensitive footprints (MSFs) and MNase-resistant footprints (MRFs) in the “leaf_diff_rep1.bdg” file. Parts with DNS values greater than 0 and Bayes values greater than 0.5 are defined as significant MSFs; parts with DNS values lower than 0 and Bayes values greater than 0.5 are defined as significant MRFs (see Note 4). The significant MSF or MRF signals within 200 bp are merged into one MSF or MRF signal. The significant MSFs are also referred to as MNase-hypersensitive (MNase HS) regions or ACRs (see Note 5).

   
   

   bedtools unionbedg \

     -i leaf_diff_rep1.bdg leaf_rep1.unionbdg.bayes \

     >leaf.diff_rep1_bayes

   
   

   cat leaf.diff_rep1_bayes | awk '$4>0 && $5>0.5' | \

     cut -f1-3 | bedtools merge -d 200 \

     >leaf.MSF_rep1_bayes_0.5_merge_200.bed

   
   

   cat leaf.diff_rep1_bayes | awk '$4<0 && $5>0.5' | \

     cut -f1-3 | bedtools merge -d 200 \

     >leaf.MRF_rep1_bayes_0.5_merge_200.bed

   
   

10. Signal visualization: The reads coverage depth of the samples obtained by different digestion treatments and the sensitivity imprint DNS values can be generated with bigwig files by deepTools and then loaded into the Integrative Genomics Viewer (IGV browser) for browsing. The MSF or MRF location files can also be loaded into the IGV browser.

    
    

    bamCoverage -b leaf_light_rep1.bam \

    --binSize 10 --smoothLength 50 \

      --normalizeUsing CPM --outFileFormat bigwig \

      -o leaf_light_rep1.bw

    bamCoverage -b leaf_heavy_rep1.bam \

    --binSize 10 --smoothLength 50 \

      --normalizeUsing CPM --outFileFormat bigwig \

      -o leaf_heavy_rep1.bw

    bamCompare -b1 leaf_light_rep1.bam -b2 leaf_heavy_rep1.bam \

      --scaleFactorsMethod None --normalizeUsing CPM --operation subtract \

      --binSize 10 --smoothLength 50 --outFileFormat bigwig \

      -o leaf_diff_rep1.bw

    
    

11. Attached R script for calculating Bayes factor: Upload the script to the same folder as “leaf_rep1.unionbdg” before use and then modify the working path, nh, nf, and other assignments according to your experiment.

    
    

    ### bayes factor caculator 1.0 ###

    
    

    setwd("data1/user/path/MNase_seq")

    
    

    M=100000

    nh=2

    nf=2

    
    

    bayes_factor_caculator <- function(old, new) {

    
    

    lkd.model1=function(y,n,lambda){ return(exp(-n*lambda+y*log(lambda))) }

    lkd.model2=function(y1,n1,y2,n2,lambda1,lambda2){ return(exp(-n1*lambda1+y1*log(lambda1)-n2*lambda2+y2*log(lambda2))) }

    
    

    BF_MC=function(a,b,y1,n1,y2,n2,M){

        lambda1=rgamma(M,a,b)

      m1=cumsum(lkd.model1(y1+y2,n1+n2,lambda1))/(1:M)

      lambda2.1=rgamma(M,a,b)

      lambda2.2=rgamma(M,a,b)

        m2=cumsum(lkd.model2(y1,n1,y2,n2,lambda2.1,lambda2.2))/(1:M)

        return(m2/m1)

      }

    
    

    con <- file(old,"rt")

    line <- readLines(con,n=1)

    
    

    while (length(line) > 0){

      line <- unlist(strsplit(line,split = "\t"))

      qh <- as.numeric(line[5])

      qf <- as.numeric(line[4])

      set.seed(1)

      bayes <- BF_MC(10,1,qh,nh,qf,nf,M)[M]

      newline=t(c(line[1],line[2],line[3],round(bayes,5)))

      write.table(newline, new, col.names = F, row.names = F, sep = '\t', quote=F, append =T)

      line <- readLines(con, n = 1)
    

    }

    
    

    close(con)

    
    

    # BF_MC(10,1,qh,nh,qf,nf,M)[M]

    # qh: total number of mapped reads for heavy-digestion

    # nh: number of replicates for heavy-digestion

    # qf: total number of mapped reads for light-digestion

    # nf: number of replicates for light-digestion

    
    

    }

    
    

    args=commandArgs(T)

    bayes_factor_caculator(args[1], paste0(args[1], '.bayes'))

Notes

1. Care should be taken when adding the filtrate to the Percoll cushion, slowly adding the filtrate along the top of the tube wall without causing violent fluctuations in the interface and mixing between liquids. When centrifuging, acceleration and deceleration must be slow. After centrifugation, a cloudy layer of liquid containing the nuclei at the interface between Percoll and the filtrate should be taken out, rather than the precipitation at the bottom of the centrifuge tube.

2. The timing of MNase and EGTA addition during the digestion reaction should be precisely controlled. Furthermore, the selection of the 100–200 bp fragment should be precise and consistent among different samples and replicates.

3. We recommend sequencing the library samples using PE150, in which case each fragment can be fully read. The sequencing depth can be determined according to the size of the genome, where the coverage of 6× is used for wheat, and the final actual performance is good. In a specific implementation, a portion of data can be preliminarily measured, and then the amount of deep sequencing data could be determined according to the effect.

4. When identifying significant MSF and MRF, if the background signal is too high, the threshold of the Bayes factor can be raised or the signals with too short length can be filtered directly to obtain more real and effective signals, which can be adjusted repeatedly in combination with the actual display of IGV browser.

5. While doing MNase-seq, we recommend performing ChIP-seq of histone modifications, such as activation modification H3K9ac, which are significant features of some ACRs and have good colocalization with ACR signals, which can be used to assist with the identification.

Acknowledgments

This protocol was applied in our open chromatin study in wheat (Kong et al., 2024), which was supported by the Central Public-interest Scientific Institution Basal Research Fund (Y2021YJ01), the National Natural Science Foundation of China (Major Program, 31991213), and the National Natural Science Foundation of China (31971882). We mainly referred to open chromatin research articles in maize (Vera et al., 2014; Rodgers-Melnick et al., 2016). In the development of this procedure, we received help from Eli Rodgers-Melnick and Daniel L. Vera. Professor Zou Cheng at Cornell University helped with our algorithms. Thanks a lot for their help!

Competing interests

There are no conflicts of interest or competing interests.

References

1. Brahma, S. and Henikoff, S. (2019). RSC-Associated Subnucleosomes Define MNase-Sensitive Promoters in Yeast. Mol. Cell. 73(2): 238-249.e3.
2. Chereji, R. V., Ocampo, J. and Clark, D. J. (2017). MNase-Sensitive Complexes in Yeast: Nucleosomes and Non-histone Barriers. Mol. Cell. 65(3): 565-577.e3.
3. Cook, A., Mieczkowski, J. and Tolstorukov, M. Y. (2017). Single-Assay Profiling of Nucleosome Occupancy and Chromatin Accessibility. Curr. Protoc. Mol. Biol. 120, 21 34 21-21 34 18.
4. Fang, Y., Wang, X., Wang, L., Pan, X., Xiao, J., Wang, X. E., Wu, Y. and Zhang, W. (2016). Functional characterization of open chromatin in bidirectional promoters of rice. Sci. Rep. 6: 32088.
5. Jordan, K. W., He, F., de Soto, M. F., Akhunova, A. and Akhunov, E. (2020). Differential chromatin accessibility landscape reveals structural and functional features of the allopolyploid wheat chromosomes. Genome. Biol. 21(1): 176.
6. Klein, D. C. and Hainer, S. J. (2020). Genomic methods in profiling DNA accessibility and factor localization. Chromosome. Res. 28(1): 69-85.
7. Lu, Z., Marand, A. P., Ricci, W. A., Ethridge, C. L., Zhang, X. and Schmitz, R. J. (2019). The prevalence, evolution and chromatin signatures of plant regulatory elements. Nat. Plants. 5(12): 1250-1259.
8. Mieczkowski, J., Cook, A., Bowman, S. K., Mueller, B., Alver, B. H., Kundu, S., Deaton, A. M., Urban, J. A., Larschan, E., Park, P. J., et al. (2016). MNase titration reveals differences between nucleosome occupancy and chromatin accessibility. Nat. Commun. 7: 11485.
9. Rodgers-Melnick, E., Vera, D. L., Bass, H. W. and Buckler, E. S. (2016). Open chromatin reveals the functional maize genome. Proc. Natl. Acad. Sci. U S A. 113(22): E3177-3184.
10. Schwartz, U., Nemeth, A., Diermeier, S., Exler, J. H., Hansch, S., Maldonado, R., Heizinger, L., Merkl, R. and Langst, G. (2019). Characterizing the nuclease accessibility of DNA in human cells to map higher order structures of chromatin. Nucleic. Acids. Res. 47: 1239-1254.
11. Tsompana, M. and Buck, M. J. (2014). Chromatin accessibility: a window into the genome. Epigenet. Chromatin. 7(1): 33.
12. Vera, D. L., Madzima, T. F., Labonne, J. D., Alam, M. P., Hoffman, G. G., Girimurugan, S. B., Zhang, J., McGinnis, K. M., Dennis, J. H. and Bass, H. W. (2014). Differential nuclease sensitivity profiling of chromatin reveals biochemical footprints coupled to gene expression and functional DNA elements in maize. Plant. Cell. 26(10): 3883-3893.
13. Yan, W., Chen, D., Schumacher, J., Durantini, D., Engelhorn, J., Chen, M., Carles, C. C. and Kaufmann, K. (2019). Dynamic control of enhancer activity drives stage-specific gene expression during flower morphogenesis. Nat. Commun. 10(1): 1705.
14. Zhao, H., Zhang, W., Zhang, T., Lin, Y., Hu, Y., Fang, C., and Jiang, J. (2020). Genome-wide MNase hypersensitivity assay unveils distinct classes of open chromatin associated with H3K27me3 and DNA methylation in Arabidopsis thaliana. Genome. Biol. 21(1): 24.

Supplementary information

1. Data and code availability: All data and code have been deposited to GitHub: https://github.com/Bio-protocol/ACRs_by_MNase-seq.