Base modification analysis in long read sequencing data using Minimod

Base modification analysis in long read sequencing data using Minimod | bioRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-M677548'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search New Results Base modification analysis in long read sequencing data using Minimod View ORCID Profile Suneth Samarasinghe , View ORCID Profile Ira Deveson , View ORCID Profile Hasindu Gamaarachchi doi: https://doi.org/10.1101/2025.07.16.665072 Suneth Samarasinghe 1 School of Computer Science and Engineering, UNSW Sydney , 2052, NSW, Australia 2 Genomics and Inherited Disease Program, Garvan Institute of Medical Research , 2010, NSW, Australia Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Suneth Samarasinghe Ira Deveson 2 Genomics and Inherited Disease Program, Garvan Institute of Medical Research , 2010, NSW, Australia 3 St Vincent’s Clinical School, UNSW Sydney , 2052, NSW, Australia Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Ira Deveson Hasindu Gamaarachchi 1 School of Computer Science and Engineering, UNSW Sydney , 2052, NSW, Australia 2 Genomics and Inherited Disease Program, Garvan Institute of Medical Research , 2010, NSW, Australia Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Hasindu Gamaarachchi For correspondence: hasindu{at}garvan.org.au Abstract Full Text Info/History Metrics Data/Code Preview PDF Abstract Recent advances in long read sequencing technologies have enabled the detection of various DNA and RNA base modifications in addition to standard nucleotide sequences. Both major vendors in this space—Oxford Nanopore Technologies (ONT) and Pacific Biosciences (PacBio)—now include base modification information in their sequencing outputs using MM/ML tags embedded in unaligned BAM files. Each vendor also provides dedicated tools for extracting and analysing these tags, such as ONT’s modkit and PacBio’s pb-CpG-tools. This work presents minimod, a new vendor-agnostic tool designed to extract and analyse any type of base modification from sequencing data generated by any platform that supports MM/ML tags. Benchmarking demonstrates that for DNA data, minimod is ~1.25× faster on a server and ~4× on a laptop compared to modkit and pb-CpG-tools. For RNA data, minimod achieves even greater speedups compared to modkit, ~12× on the server and ~55× on the laptop. Minimod is a free, open-source application written in C and is available at https://github.com/warp9seq/minimod . 1. Introduction Base modifications are considered epigenetic as they involve chemical changes to nucleotide molecules that do not alter the underlying DNA sequence. These modifications are known to play a role in regulating gene expression across different cell types and have been studied in the context of disease, ageing and forensics [ 1 , 2 , 3 , 4 , 5 , 6 ]. The reversible nature of epigenetic modifications offers a promising pathway to effective therapeutics [ 7 ]. The most prevalent epigenetic modification in the eukaryotic genome is DNA methylation, where a methyl group attaches to the C5 position of the Cytosine to form 5-methylcytosine (5mC). DNA methylation is predominantly located in CpG dinucleotides, while found less frequently in non-CpGs [ 8 , 9 ]. Methylation levels in the promoter region of a gene can regulate its expression by switching the gene on or off [ 8 ]. Silencing of tumour suppressor genes is known to increase an individual’s risk of developing certain cancers [ 10 ]. Bisulphite sequencing, which relies on short-read sequencing, is considered the gold standard for identifying DNA methylation [ 11 ]. However, bisulphite conversion can cause complications due to extensive fragmentation of DNA during deamination, and reliance on short reads limits detection to only short-range methylation patterns [ 12 ]. In contrast, long-read sequencing technologies such as single-molecule real-time (SMRT) sequencing from Pacific Biosciences (PacBio) and nanopore sequencing from Oxford Nanopore Technologies can sequence native DNA molecules and generate much longer reads, enabling the detection of long-range methylation patterns [ 12 , 13 ]. In addition to detecting DNA methylation, long-read sequencing technologies can identify various other modifications in both DNA and RNA. There are around 58 known DNA modification types [ 14 ] and over 170 types for RNA [ 15 ]. Currently, PacBio can detect three DNA (4mC, 5mC and 6mA) and two RNA (m6A, m4C) modifications, while ONT can detect four DNA (4mC, 5mC, 5hmC, 6mA) and four RNA (m5C, m6A, Inosine, pseU) modifications [ 16 , 17 , 18 , 19 ]. This number is expected to grow as long-read sequencing technologies theoretically have the capability to detect any type of molecular modification. Both major long-read vendors, ONT and PacBio, now produce unaligned S/BAM files that encode base modifications using auxiliary MM and ML tags, as defined by the official SAM tags specification. These tags can be retained during read alignment to a reference genome and subsequently used in downstream analyses, such as computing base modification frequencies. For such analyses, vendor-specific tools can be used, for example, PacBio’s pb-CpG-tools and ONT’s modkit. Although both tools are open-source, they are distributed under non-standard licenses: pb-cpg-tools under the Pacific Biosciences Software License Agreement, and modkit under the Oxford Nanopore Technologies PLC Public License 1.0. Such non-standard licenses often impose restrictions. For example, the Pacific Biosciences Software License Agreement limits the use of its software to data generated exclusively on PacBio instruments, while the Oxford Nanopore Technologies PLC Public License restricts commercial use. Here, we introduce minimod, a fully open-source (MIT-licensed), vendor-independent software tool capable of processing any type of base modification in any sequence context. Minimod supports all platforms that encode modification information using MM/ML tags, including those from ONT and PacBio. Minimod supports both DNA and RNA modifications. Our benchmarking demonstrates that minimod is ~1.25× faster on a server and ~4× faster on a laptop compared to both modkit and pb-CpG-tools when processing DNA data. For RNA data, minimod delivers a significant performance improvement, achieving ~12× faster execution on the server and ~55× on the laptop compared to modkit. 2. Methods and implementation Minimod is implemented using the C programming language and currently has two subtools: view for extracting modifications along with modification probabilities; and, freq for computing modification frequencies. Both subtools take a BAM file containing modification tags and the reference genome as input. 2.1. Minimod software architecture Minimod is designed and implemented for high performance and efficient resource utilisation. Minimod supports parallel processing through multithreading and pipeline stages implemented using POSIX threads. For a small memory footprint and overhead, hash tables are implemented using the lightweight khash from the klib library. htslib is used to load BAM files and to retrieve SAM auxiliary tags. 2.1.1. Data structure initialisation In the initialisation phase, command line arguments are parsed, validated and stored to be used in later steps. All possible data structures are allocated and initialised, including those used within threads. Such allocated data structures are reused later whenever possible to minimise the time overhead of dynamic memory allocations during runtime. After memory allocation, the reference genome/transciptome is loaded into memory. When loading the reference, we perform the Knuth-Morris-Pratt pattern searching algorithm on each reference sequence to identify the contexts requested by the user and mark them in a genome-wide array. This array is used later to identify if a modification lies within a requested context ( section 2.2.4 ). The reference sequence is then unloaded to save memory. 2.1.2. Parallel processing After initialisation, the processing of BAM records is performed in batches (batch size can be user-specified, see section 2.2.3 ) using three interleaved pipeline stages. Both subtools have load and process as the first two pipeline stages. The third pipeline stage in the view subtool is output , while in the freq subtool it is merge . view subtool: In the load pipeline stage, a batch of BAM records is loaded into memory from the disk. For decompressing BAM records, multiple threads are used. In the process pipeline stage, multiple threads are launched to compute base alignment and extract modifications from MM and ML tags (see section 2.1.3 ). These decoded modifications are stored in per-record data hash tables after checking if they lie within a requested context using the context array initialised in section 2.1.1 . Finally, the output pipeline stage prints the extracted modifications to a given destination. freq subtool: The load and process stages are identical to those in the view subtool above. When computing modification frequencies, we need to aggregate modifications based on contig name, position, modification type and strand (and haplotype if specified, see section 2.2.6 ), and compute a frequency value for each entry. This is achieved by merging modifications from per-record hash tables into a global hash table in the merge pipeline stage. This is performed by a single thread to avoid the need for locking. Before updating an entry in the hash table, a modification call is categorised based on the modification probability values as discussed in section 2.2.5 . Fig. 1A shows how each batch is progressed through the three pipeline stages of the freq subtool. The load stage of the second batch starts together with the start of the process stage of the first batch. Finally, when the last batch is merged, a single-threaded output stage writes the modifications along with frequency values. Download figure Open in new tab Fig. 1: A) Execution block diagram of minimod freq with interleaved stages, load (load a batch of records from input BAM), process (extract modifications from the batch), merge (merge modifications into a genome-wide hash table), followed by the final non-interleaved output stage. B) Classification of the modifications based on the modification probability threshold. C) Correlation of methylation frequency results for the ONT DNA dataset. Panel 1 compares modkit with bisulphite, panel 2 compares minimod with bisulphite, and panel 3 compares minimod with modkit. D) Correlation of methylation frequency results for the PacBio DNA dataset. Panel 1 compares pb-CpG-tools with bisulphite, panel 2 compares minimod with bisulphite, and panel 3 compares minimod vs pb-CpG-tools. E) Correlation of methylation frequency results for the ONT RNA dataset with minimod vs modkit. F) Execution time of modification frequency computation using minimod, modkit and pb-CpG-tools on a server and laptop computer. 2.1.3. BAM record processing When computing base alignment, the aligned reference position is calculated by iterating through the CIGAR string of a BAM record. The read-to-reference alignment is stored in an array indexed by read position. When extracting modifications from MM and ML tags, minimod iterates through and decodes the MM tag, which is a string that indicates the canonical base, strand, type of modification, and the location of modification as a list of delta-encoded positions. Similarly, minimod decodes the ML tag, which is a byte array containing modification likelihoods (probabilities) for each modification reported in the MM tag. The MM/ML tags are parsed using our efficient implementation described in realfreq [ 20 ]. 2.2. Features and usage The only external dependencies of minimod are htslib and zlib ; thus, minimod can be compiled with minimal effort. Also, binaries for x86 64 are provided for further convenience of the user. Usage of currently available subtools and their features are as follows. 2.2.1. view subtool The view subtool extracts modifications from auxiliary SAM tags and writes to the output in TSV format. A simple usage command is as follows, which writes all 5mC modifications found in the CpG sites (default modification code “ m ” and context “ CG ”, explained in section 2.2.4 ) to mods.tsv file. Download figure Open in new tab 2.2.2. freq subtool The freq subtool extracts modifications and aggregates them based on contig, position, and modification type to compute frequency values, followed by writing the output in TSV format (default) or BED format (if specified with -b flag). A simple example command is given below. Download figure Open in new tab This writes the modification frequencies for 5mC found in CpG sites (default modification code “ m ” and context “ CG ”, explained in section 2.2.4 ), considering 0.8 as the default modification threshold (explained in section 2.2.5 ), to the modfreqs.tsv file. 2.2.3. Batch size and threads As discussed in section 2 , minimod processes batches of BAM records, with the load and process pipeline stages utilising multithreading. The batch size and the number of threads are user-configurable options, allowing adjustment to suit the available system memory and number of CPUs. The -K option sets the maximum number of BAM records per batch, -B option sets the maximum number of bases per batch and -t sets the number of threads. 2.2.4. Filter by modification code and context For both subtools, minimod can output a specific type of modification(s) observed in a specific context in the reference using the -c option. It takes an argument in the format of x or x [ y ]. Here x is a single character representing the modification type/code specified in the official SAM tags specification (e.g., m, h ) or a ChEBI number for non-standard modifications. y is a multi-character context (e.g., CG, A , or * representing all contexts). If only x is specified, the default modification code and context m and CG are assumed for y . The following are a few example arguments for the -c option. a [ A ]: m6A modifications of all A bases h [ CG ]: 5hmC modifications in CG context (CpG sites) m : 5mC modifications in the default CG context a [ * ]: m6A modifications in all contexts 17802 [ T ]: pseU (ChEBI: 17802) modifications in T context Minimod uses the reference to precisely identify the context when filtering modifications based on context. This is because solely relying on the sequence in the BAM record can be erroneous if an actual variant is present. 2.2.5. Filter by modification threshold The modification threshold value is only applicable to the freq subtool and can be specified by the -m option. If not specified, the default threshold value of 0.8 is assumed. It is used to categorise modification calls as modified, unmodified, or ambiguous, based on the modification probability as follows. It is visually represented in Fig. 1B . Download figure Open in new tab 2.2.6. Haplotypes and modifications in insertion Minimod can output the haplotype information in a separate column for both view and freq subtools by specifying the --haplotype flag. When enabled with freq , in addition to separate entries for each haplotype, an additional entry will be printed considering all haplotypes indicated by a * in the haplotype column. Furthermore, modifications within inserted regions can be included in the output of both view and freq subtools, by specifying the --insertions flag. This will add a new column, ins offset , with the position of the modified base within the inserted region. However, the --haplotype and --insertions flags are only applicable when the output is in TSV format. 3. Evaluations and results 3.1. Datasets and computer Systems To evaluate DNA modifications, we used two HG002 human genome datasets, one dataset sequenced on an ONT PromethION (henceforth referred to as ONT DNA ) and the other on a PacBio Revio ( PacBio DNA ). To evaluate RNA modifications, we used a Universal Human Reference RNA sample sequenced using an ONT PromethION ( ONT RNA ). See Supplementary Table S1 for details of the datasets. HG002 Bisulphite data was downloaded from the publicly available ONT open-data AWS repository (s3://ont-opendata/gm24385 mod 2021.09/bisulphite/cpg). We conducted benchmarks on a laptop and a high-performance server to evaluate minimod and compare its performance with ONT’s modkit and PacBio’s pb-CpG-tools. The laptop computer used for the evaluation was equipped with an Intel i71370P CPU and 32 GB of RAM. The server computer was equipped with two Intel Xeon Gold 6154 CPUs and 376 GB of RAM. Both systems were running Ubuntu 22.04 as the operating system. See Supplementary Table S2 for more details of the systems. 3.2. Evaluation We evaluated minimod in terms of accuracy and execution performance. The ONT datasets were first converted from POD5 to BLOW5 using blue-crab [ 21 , 22 ]. The ONT DNA dataset was basecalled using buttery-eel [ 23 ] with 5hmc 5mc modification calling in CG context. The ONT RNA dataset was basecalled using slow-dorado with m6A DRACH modification calling. This produced unaligned SAM/BAM output with MM/ML tags. The PacBio DNA dataset was already in the unaligned SAM/BAM format with MM/ML tags. The unaligned SAM/BAM files were aligned using minimap2 [ 24 ] (with -Y -y options to retain MM/ML tags) and were sorted using samtools [ 25 ] to produce BAM files with MM/ML tags. For DNA data, the hg38 reference was used for the alignment. For RNA data, the Gencode v40 transcriptome was used. To measure accuracy, modification frequency was obtained by running minimod v0.4.0 freq on all three datasets (sorted BAM with MM/ML tags), modkit 0.5.0 on two ONT datasets, and pb-CpG-tools 2.3.2 on the PacBio dataset. Methylation (5mC) modification was considered for the two DNA datasets and N6-methyladenosine (m6A) for the RNA dataset. Comparison of methylation frequency results and plotting the correlation were performed by adapting the scripts compare methylation . py and plot methylation . R from the f5c/nanopolish repository [ 26 , 27 ] to support the minimod TSV format. To measure execution performance, each experiment was performed five times on both laptop and server computers. We used /usr/bin/time with -v flag when running each tool and took elapsed (wall clock) time as the runtime metric for performance evaluation. All software was executed with 32 threads on the server and with 8 threads on the laptop. Software versions and commands that we executed are in Supplementary Note 1. 3.3. Accuracy comparison For the ONT DNA dataset, both modkit and minimod show similar Pearson correlation coefficients of ~0.95 when compared against the bisulphite truthset ( Fig. 1C , first and second panels). However, minimod has ~15000 more data points in common with the truthset than modkit has with the truthset. When compared with each other, minimod vs modkit has a correlation of ~0.99 ( Fig. 1C , third panel). For the PacBio DNA dataset, methylation results from pb-CpG-tools show a slightly higher correlation (~0.87) than minimod (~0.86), as shown in Fig. 1D (first and second panels). However, minimod has ~359,000 more data points in common with the truthset than pb-CpG-tools has with the truthset, which may compensate for the difference in correlation. When compared with each other, minimod vs pb-CpG-tools shows a correlation of ~0.99 ( Fig. 1D , third panel). For the ONT RNA dataset, m6A modification results between modkit vs minimod show a correlation of ~0.99 ( Fig. 1E ). For RNA modification data, we did not have access to a ground truth set, unlike the bisulphite-based reference available for DNA The differences in the number of data points reported by each tool and the minor differences in correlation could be due to different filtering methods and threshold values used in other tools compared to minimod ( section 2.2.5 ). 3.4. Performance comparison Minimod is faster than both modkit and pb-CpG-tools for all three datasets on both the server ( Fig. 1F left panel) and the laptop ( Fig. 1F right panel). On the server, minimod is ~1.2× faster than modkit when processing the ONT DNA dataset, ~12.6× faster than modkit when processing ONT RNA dataset, and ~1.3× faster than pb-CpG-tools when processing the PacBio DNA dataset. On the laptop, minimod is ~4.1× faster than modkit when processing the ONT DNA dataset, ~54.8× faster than modkit when processing the ONT RNA dataset, and ~4.0× faster than pb-CpG-tools when processing the PacBio DNA dataset. 4. Conflict of interest S.S., H.G. and I.W.D. have received travel and accommodation expenses from Oxford Nanopore Technologies. I.W.D. manages a fee-for-service sequencing facility at the Garvan Institute of Medical Research and is a customer of Oxford Nanopore Technologies, but has no further financial relationship. 6. Data and code availability The source code for minimod is open-source at https://github.com/warp9seq/minimod under the MIT license. Minimod v0.4.0 used for the benchmarks is also at doi:10.5281/zenodo.15876999. Unaligned and aligned BAM files with modification calls are available at do:10.5061/dryad.c59zw3rm7. Raw nanopore signal data is available at ENA (HG002 sample: ERR12997168 and UHRR sample: ERR12997170). Supplementary Material Supplementary Note 1: Commands and versions of software used View this table: View inline View popup Download powerpoint Supplementary Table S1: Datasets used for testing View this table: View inline View popup Download powerpoint Supplementary Table S2: Computer systems used for testing Download figure Open in new tab Versions buttery-eel: 0.6.0 ont-dorado-server: 7.4.12 slow5-dorado: 0.8.3 modkit: 0.5.0 minimod: 0.4.0 pb-CpG-tools: 2.3.2 samtools: 1.21 minimap2: 2.28 5. Acknowledgement We thank our colleagues James Ferguson, Leah Kemp, Marjan Naeini, Jillian Hammond, Igor Stevanovski, Tonia Russell and Andre Martins Reis at the Garvan Institute for their assistance at different stages of this work. We acknowledge the following funding support: Australian Research Council DECRA Fellowship DE230100178 (to H.G. and PhD scholarship for S.S.). Australian Medical Research Futures Fund grants MRF2016008 and MRF2023126 (to I.W.D.). Funder Information Declared Australian Research Council , DE230100178 National Health and Medical Research Council , MRF2016008 , MRF2023126 Footnotes https://github.com/warp9seq/minimod http://doi.org/10.5061/dryad.c59zw3rm7 References 1. ↵ John Newell-Price , Adrian JL Clark , and Peter King . Dna methylation and silencing of gene expression . Trends in Endocrinology & Metabolism , 11 ( 4 ): 142 – 148 , 2000 . OpenUrl CrossRef PubMed 2. ↵ Yongkyu Park and Mitzi I Kuroda . Epigenetic aspects of x-chromosome dosage compensation . Science , 293 ( 5532 ): 1083 – 1085 , 2001 . OpenUrl Abstract / FREE Full Text 3. ↵ Andrew P Feinberg and Bert Vogelstein . Hypomethylation distinguishes genes of some human cancers from their normal counterparts . Nature , 301 ( 5895 ): 89 – 92 , 1983 . OpenUrl CrossRef PubMed Web of Science 4. ↵ Miguel A Gama-Sosa , Valerie A Slagel , Ronald W Trewyn , Ronald Oxenhandler , Kenneth C Kuo , Charles W Gehrke , and Melanie Ehrlich . The 5-methylcytosine content of dna from human tumors . Nucleic acids research , 11 ( 19 ): 6883 – 6894 , 1983 . OpenUrl CrossRef PubMed Web of Science 5. ↵ Marco P Boks , Eske M Derks , Daniel J Weisenberger , Erik Strengman , Esther Janson , Iris E Sommer , Rene S Kahn , and Roel A Ophoff . The relationship of dna methylation with age, gender and genotype in twins and healthy controls . PloS one , 4 ( 8 ): e6767 , 2009 . OpenUrl CrossRef PubMed 6. ↵ Dan Frumkin , Adam Wasserstrom , Bruce Budowle , and Ariane Davidson . Dna methylation-based forensic tissue identification . Forensic Science International: Genetics , 5 ( 5 ): 517 – 524 , 2011 . OpenUrl PubMed 7. ↵ Masoumeh Fardi , Saeed Solali , and Majid Farshdousti Hagh . Epigenetic mechanisms as a new approach in cancer treatment: An updated review . Genes & Diseases , 5 ( 4 ): 304 – 311 , December 2018 . OpenUrl PubMed 8. ↵ Lisa D Moore , Thuc Le , and Guoping Fan . Dna methylation and its basic function . Neuropsychopharmacology , 38 ( 1 ): 23 – 38 , 2013 . OpenUrl CrossRef PubMed Web of Science 9. ↵ Wai-Shin Yong , Fei-Man Hsu , and Pao-Yang Chen . Profiling genome-wide dna methylation . Epigenetics & chromatin , 9 ( 1 ): 26 , 2016 . OpenUrl CrossRef PubMed 10. ↵ Franziska Geissler , Ksenija Nesic , Olga Kondrashova , Alexander Dobrovic , Elizabeth M Swisher , Clare L Scott , and Matthew J. Wakefield . The role of aberrant dna methylation in cancer initiation and clinical impacts . Therapeutic Advances in Medical Oncology , 16 : 17588359231220511 , 2024 . 11. ↵ Russell P Darst , Carolina E Pardo , Lingbao Ai , Kevin D Brown , and Michael P Kladde . Bisulfite sequencing of dna . Current protocols in molecular biology , 91 ( 1 ): 7 – 9 , 2010 . OpenUrl 12. ↵ Yang Liu , Wojciech Rosikiewicz , Ziwei Pan , Nathaniel Jillette , Ping Wang , Aziz Taghbalout , Jonathan Foox , Christopher Mason , Martin Carroll , Albert Cheng , et al. Dna methylation-calling tools for oxford nanopore sequencing: a survey and human epigenome-wide evaluation . Genome biology , 22 ( 1 ): 295 , 2021 . OpenUrl CrossRef PubMed 13. ↵ Yilei Fu , Winston Timp , and Fritz J Sedlazeck . Computational analysis of dna methylation from long-read sequencing . Nature Reviews Genetics, pages 1 – 15 , 2025 . 14. ↵ Ankur Jai Sood , Coby Viner , and Michael M Hoffman . Dnamod: the dna modification database . Journal of cheminformatics , 11 ( 1 ): 30 , 2019 . OpenUrl CrossRef PubMed 15. ↵ Andrea Cappannini , Angana Ray , Elz?bieta Purta , Sunandan Mukherjee , Pietro Boccaletto , S Naeim Moafinejad , Antony Lechner , Charles Barchet , Bruno P Klaholz , Filip Stefaniak , et al. Modomics: a database of rna modifications and related information. 2023 update . Nucleic Acids Research , 52 ( D1 ): D239 – D244 , 2024 . OpenUrl CrossRef PubMed 16. ↵ Anthony Rhoads and Kin Fai Au . Pacbio sequencing and its applications . Genomics, proteomics & bioinformatics , 13 ( 5 ): 278 – 289 , 2015 . OpenUrl CrossRef PubMed 17. ↵ Ying Chen , Nadia M Davidson , Yuk Kei Wan , Fei Yao , Yan Su , Hasindu Gamaarachchi , Andre Sim , Harshil Patel , Hwee Meng Low , Christopher Hendra , et al. A systematic benchmark of nanopore long-read rna sequencing for transcript-level analysis in human cell lines . Nature methods, pages 1 – 12 , 2025 . 18. ↵ Alan Tourancheau , Edward A Mead , Xue-Song Zhang , and Gang Fang . Discovering multiple types of dna methylation from bacteria and microbiome using nanopore sequencing . Nature methods , 18 ( 5 ): 491 – 498 , 2021 . OpenUrl PubMed 19. ↵ William Stephenson , Roham Razaghi , Steven Busan , Kevin M Weeks , Winston Timp , and Peter Smibert . Direct detection of rna modifications and structure using single-molecule nanopore sequencing . Cell genomics , 2 ( 2 ), 2022 . 20. ↵ Suneth Samarasinghe , Ira Deveson , and Hasindu Gamaarachchi . Realfreq: real-time base modification analysis for nanopore sequencing . Bioinformatics , 41 ( 4 ): btaf151 , 2025 . OpenUrl PubMed 21. ↵ Hasindu Gamaarachchi , Hiruna Samarakoon , Sasha P Jenner , James M Ferguson , Timothy G Amos , Jillian M Hammond , Hassaan Saadat , Martin A Smith , Sri Parameswaran , and Ira W Deveson . Fast nanopore sequencing data analysis with slow5 . Nature biotechnology , 40 ( 7 ): 1026 – 1029 , 2022 . OpenUrl CrossRef PubMed 22. ↵ Hiruna Samarakoon , James M Ferguson , Sasha P Jenner , Timothy G Amos , Sri Parameswaran , Hasindu Gamaarachchi , and Ira W Deveson . Flexible and efficient handling of nanopore sequencing signal data with slow5tools . Genome Biology , 24 ( 1 ): 69 , 2023 . OpenUrl CrossRef PubMed 23. ↵ Hiruna Samarakoon , James M Ferguson , Hasindu Gamaarachchi , and Ira W Deveson . Accelerated nanopore basecalling with slow5 data format . Bioinformatics , 39 ( 6 ): btad352 , 2023 . OpenUrl CrossRef PubMed 24. ↵ Heng Li . Minimap2: pairwise alignment for nucleotide sequences . Bioinformatics , 34 ( 18 ): 3094 – 3100 , 2018 . OpenUrl CrossRef PubMed 25. ↵ Petr Danecek , James K Bonfield , Jennifer Liddle , John Marshall , Valeriu Ohan , Martin O Pollard , Andrew Whitwham , Thomas Keane , Shane A McCarthy , Robert M Davies , et al. Twelve years of samtools and bcftools . Gigascience , 10 ( 2 ): giab008 , 2021 . OpenUrl CrossRef PubMed 26. ↵ Hasindu Gamaarachchi , Chun Wai Lam , Gihan Jayatilaka , Hiruna Samarakoon , Jared T Simpson , Martin A Smith , and Sri Parameswaran . Gpu accelerated adaptive banded event alignment for rapid comparative nanopore signal analysis . BMC bioinformatics , 21 ( 1 ): 343 , 2020 . OpenUrl CrossRef PubMed 27. ↵ Jared T Simpson , Rachael E Workman , PC Zuzarte , Matei David , LJ Dursi , and Winston Timp . Detecting dna cytosine methylation using nanopore sequencing . Nature methods , 14 ( 4 ): 407 – 410 , 2017 . OpenUrl CrossRef PubMed View the discussion thread. Back to top Previous Next Posted July 21, 2025. Download PDF Data/Code Email Thank you for your interest in spreading the word about bioRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Base modification analysis in long read sequencing data using Minimod Message Subject (Your Name) has forwarded a page to you from bioRxiv Message Body (Your Name) thought you would like to see this page from the bioRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Base modification analysis in long read sequencing data using Minimod Suneth Samarasinghe , Ira Deveson , Hasindu Gamaarachchi bioRxiv 2025.07.16.665072; doi: https://doi.org/10.1101/2025.07.16.665072 Share This Article: Copy Citation Tools Base modification analysis in long read sequencing data using Minimod Suneth Samarasinghe , Ira Deveson , Hasindu Gamaarachchi bioRxiv 2025.07.16.665072; doi: https://doi.org/10.1101/2025.07.16.665072 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Bioinformatics Subject Areas All Articles Animal Behavior and Cognition (7633) Biochemistry (17680) Bioengineering (13889) Bioinformatics (41928) Biophysics (21445) Cancer Biology (18585) Cell Biology (25491) Clinical Trials (138) Developmental Biology (13374) Ecology (19897) Epidemiology (2067) Evolutionary Biology (24308) Genetics (15606) Genomics (22496) Immunology (17736) Microbiology (40385) Molecular Biology (17175) Neuroscience (88583) Paleontology (666) Pathology (2830) Pharmacology and Toxicology (4822) Physiology (7641) Plant Biology (15149) Scientific Communication and Education (2045) Synthetic Biology (4293) Systems Biology (9822) Zoology (2271)