Novel nanopore sequencing method for determining Human Papillomavirus integrations in tumors without the need for whole genome sequencing

43 Human papillomaviral (HPV) integrations into host human genome, a key event in cervical 44 carcinogenesis, are currently mapped through laborious and expensive sequencing 45 methodologies. We developed and validated a novel library preparation strategy for nanopore 46 sequencing to generate long targeted reads with HPV and human chimeric sequences. Using this 47 strategy, we validated known HPV integrations in HeLa (HPV18) and SiHa (HPV16) cell lines. 48 We also mapped integration sites in five HPV+ cervical cancer patients, which were confirmed 49 by whole genome and Sanger sequencing. Our nanopore-based method provides a precise and 50 efficient strategy to capture HPV integrations crucial for understanding tumorigenesis. 51

HPV integration, Cervical Cancer, nanopore sequencing 52

53 Human papillomavirus (HPV) infections are associated with oropharyngeal, anogenital, genital, 54 head and neck, and cervical cancers (1). Cervical cancer is the fourth most common cancer in 55 women worldwide, which according to GLOBOCAN 2022, accounted for more than 662,301 56 new cases reported and 348,874 deaths (2,3). Persistent HPV infection is the primary driver of 57 more than 99% of cervical cancers (4), 70% of which are caused by high-risk HPV types 16 and 58 18 (5). HPV is a non-enveloped double-stranded DNA virus belonging to the Papillomaviridae 59 family. It infects the basal cells of the cervical epithelium and replicates as these cells divide and 60 differentiate to form the upper epithelial layers, eventually being shed from the top layer as 61 progeny virions that initiate reinfection (6). HPV induces carcinogenesis by inhibiting the host 62 tumor suppressor proteins, p53 and retinoblastoma protein (Rb) using the viral proteins- E6 and 63 E7 respectively (7). Integration of the HPV genome into a host chromosome is also a crucial step 64 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint in HPV-induced carcinogenesis and accompanies the formation of invasive cervical cancer that 65 breaks through the basement membrane. Integration sites are more likely to be enriched at 66 common fragile sites and open chromatin regions in the human genome (8–10). Large-scale 67 genomic analyses have also uncovered hotspots of recurrent HPV integration with significant 68 enrichment for sequences with microhomology between the human and HPV genomes at the 69 breakpoints (11). Tumors with HPV integration show upregulation of viral oncogenes E6 and 70 E7, as well as host genes at or near the site of integration (12). Recent studies have observed that 71 genomic regions around HPV integrations are enriched with Gene Ontology (GO) terms for 72 DNA repair, fueling the hypothesis that integration events in the vicinity of DNA repair and 73 tumor suppressor genes could lead to increased genomic instability (13). Moreover, HPV 74 integration has been linked with changes in host chromatin structure and, consequently, in gene 75 regulation, including long-range interactions (10). 76 Numerous such studies highlighting the cis and trans effects of HPV integration in inducing and 77 promoting cervical carcinogenesis have underscored the importance of HPV integration 78 detection in cancer research, leading to a surge in developing technologies to study HPV 79 integrations (9). The Cancer Genome Atlas characterized 228 primary cervical cancers and 80 reported HPV integrations in all HPV18-positive samples and 76% of HPV16-positive samples 81 (10). Campitelli et al. used these cell-viral junctions as a biomarker in circulating tumor DNA for 82 analyzing a series of serum samples obtained from cervical cancer patients and reported that 83 HPV integration can be used as a biomarker for the detection of minimal residual disease and 84 subclinical relapse in HPV-associated cancers (14). Therefore, detecting HPV integration status 85 and identifying the integration locus are crucial steps not only for understanding the role of HPV 86 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint in cervical carcinogenesis, but also for clinical diagnosis and treatment monitoring in cervical 87 cancer patients (9). 88 Various techniques have been employed to determine HPV integration sites in the human 89 genome, such as detection of integrated papillomavirus sequences by ligation-mediated PCR 90 (DIPS-PCR) (15), Restriction Site PCR (RS-PCR) (16), amplification of papillomavirus 91 oncogene transcripts (APOT) and next-generation sequencing (NGS). RNA-based assays for 92 detection of HPV integration sites have substantial biological and technological constraints 93 because of the lower stability of mRNA in biopsy samples (15). Though assays like DIPS-PCR 94 are dependent on DNA, the potential for detecting new integration sites is limited (9). Recently, 95 NGS has been widely used for the genome-wide characterization of HPV integrations in cervical 96 cancer (17). However, the method requires whole genome sequencing at high coverage (>30X) 97 (18), and involves multiple steps (19). Considering the complexity of the procedure and the 98 resulting data (19), this method, though accurate, is not suitable for clinical use (9). Further, the 99 cost of reagents used for these assays can be a challenge for labs with less resources, space, and 100 medium to low sample throughput (20). With lower reagent costs and a shorter sequencing time, 101 portable Nanopore sequencers may provide greater flexibility in this aspect (20). 102 Nanopore sequencing technology is the fourth generation of sequencing technology. It is 103 portable, provides real-time data and enables rapid sequencing in clinical settings (21). In this 104 study, as a proof of concept, we developed a novel enrichment strategy to capture HPV-human 105 integration breakpoints using nanopore sequencing. We validated this technology by mapping 106 known HPV integrations from cell lines and patient samples. Furthermore, by demonstrating 107 expression changes of cancer-associated genes near the integration sites in the patient samples, 108 we highlight the functional significance of mapping HPV integration events. 109 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint

110 Cell Culture 111 Cervical cancer cell lines SiHa (HPV-16 positive), and HeLa (HPV-18 positive), procured from 112 the National Centre for Cell Sciences (NCCS), Pune, India, were cultured in Dulbecco’s 113 modified Eagle’s medium (DMEM) (HiMedia Laboratories Pvt. Ltd., Mumbai, India) 114 supplemented with 10% fetal bovine serum (HiMedia Laboratories Pvt. Ltd., Mumbai, India) and 115 1% antibiotic/antimycotic solution (HiMedia Laboratories Pvt. Ltd., Mumbai, India). 116 117 Patient Recruitment And Sample Collection 118 Ethical approval for the study was obtained from the Institutional Ethics Committee, Kasturba 119 Medical College Manipal, Manipal Academy of Higher Education, Manipal, India Fresh tumor 120 biopsies from five patients with cervical cancer, admitted to Kasturba Medical College, Manipal, 121 were collected after obtaining informed consent and prior initiating cancer therapy. Patient 122 clinical parameters like age, histological grade, comorbidity status, etc., were obtained from the 123 medical records and summarized in Table 1. Biopsy samples were snap-frozen in liquid nitrogen 124 or dry ice ethanol bath and stored at −80 °C until further use. 125 126 DNA Isolation, HPV Detection and Genotyping 127 DNA was extracted from tumor biopsy samples using a DNeasy Blood and Tissue kit (Qiagen, 128 Hilden, Germany) following the manufacturer’s protocol. The DNA concentration of each 129 sample was measured using Qubit® fluorometer 3.0 (Thermo Fisher Scientific Inc., USA) using 130 the Qubit™ dsDNA BR Assay Kit. HPV detection by PCR was performed using HPV universal 131 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint primers GP5+ and GP6+. PCR reaction comprised of 1x GoTaq® Green Master Mix (Promega 132 Corp., WI, USA), 0.5 μ M of GP5+ and GP6+ primers and 50 ng of genomic DNA. Thermal 133 cycling was performed using SimpliAmp™ Thermal Cycler (Thermo Fisher Scientific Inc., 134 USA) with the following conditions: Initial denaturation at 95°C for 2 min, 95°C for 30 s, 51°C 135 for 30 s, 72°C for 30s, 35 cycles and final extension at 72°C for 5 min for 1 cycle. 136 137 Whole Genome Sequencing (WGS) and analysis 138 Genomic DNA from tumor samples was sequenced using the Illumina Novaseq/ NextSeq 139 sequencer as per the manufacturer’s instructions (Illumina, San Diego, California), with 2 × 150-140 bp paired-end reads and a minimum coverage of approximately 30X. WGS was performed at 141 Medgenome laboratories based in Bangalore, India. FASTQ files from the paired-end WGS were 142 passed through quality control and adapter trimming using the Trim Galore wrapper over 143 Cutadapt and FastQC (22). To detect viral integration, the trimmed FASTQ files for each sample 144 were processed through the ViFi algorithm using hg38 as the human reference genome, along 145 with HPV genomes (23). ViFi employs phylogenetic methods in conjunction with reference-146 based read mapping to accurately identify integration events, even for novel viral strains. This 147 analysis revealed samples with viral integration, providing details on the viral strain, the human 148 chromosome and positional range of integration, as well as the number and identities of read 149 pairs supporting the integration. Supporting reads include pairs in which one read maps to the 150 human reference genome and the other to a viral reference genome, as well as split reads—a type 151 of chimeric read that spans the integration site, mapping partially to the human and viral 152

genomes. Split read sequences were fetched from the BAM file outputs from ViFi and 153 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint verified by visualizing the alignments in the Integrative Genomics Viewer (IGV). Precise 154 integration points were then identified from the split read sequences through BLAST analysis 155 against the human reference genome and the detected viral strain. 156 157 Single primer PCR extension 158 We developed a novel single primer extension strategy for specifically targeting HPV fragments. 159 The schematic of the single primer extension based nanopore library preparation is depicted in 160 Figure 1. This method for detecting HPV-human breakpoints involves a pool of primers flanking 161 which cover the entire HPV genome. Two sets of primer pools were designed which amplify 162 specific HPV genomic regions in forward and reverse directions respectively. The elongation of 163 a single primer can capture HPV fragments and adjacent human genomic sequences during 164 primer extension. The homopolymer tailing (C-Tailing) step followed by the single primer 165 extension produces double-stranded DNA fragments and inhibits potential PCR-generated 166 artefacts caused by errors in polymerase incorporation. This method can also detect novel 167 integration breakpoints, unlike the traditional HPV detection PCRs which can only amplify a 168 certain number of integration sites based on specific genes of HPV (E1, E2 OR L1) (15). Our 169

can also be instrumental in assessing HPV integration sites in highly fragmented or 170 damaged DNA such as formalin fixed paraffin embedded (FFPE) tissue DNA, as it does not 171 require a predefined amplicon length, unlike conventional PCR methods. The final round of 172 nested amplification using nanopore adapter specific primers makes the assay more specific by 173 enriching the amplicons obtained in the first round of primer extension. The use of nanopore 174 specific adapter helps in barcoding, thereby multiplexing samples, which further reduces the cost 175 of sequencing. The use of this library preparation technique for nanopore sequencing reduces the 176 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint need for whole genome sequencing for detecting HPV-human integration events while 177 increasing the efficacy of detecting HPV-human hybrid fragments even at very low copies. 178 Primer Design 179 HPV16 and 18 whole genome references were obtained from the Papilloma Virus Episteme 180 (PaVE) database (pave.niaid.nih.gov) (24). Primers were designed using PRIMER3 software 181 such that they tile the whole HPV genome. These primers are listed in Supplementary Table 1. 182 Tiled primer design was accomplished with 24 primers (10 in forward primer pool and 14 in 183 reverse primer pool) each spanning an average of 500bp-1kb per genome of HPV16, and with 17 184 primers (9 in forward pool and 8 in reverse pool) each spanning an average of 1-1.5kb for 185 HPV18. 186 First round PCR Extension 187 First round of single primer extension was performed using 50 pmoles of HPV-specific primers 188 with nanopore-specific adapters for barcoding, 500ng of genomic DNA extracted from the 189 respective cell line or tumor samples, 10μ l 5x PrimeSTAR GXL Buffer, 4μ l dNTP mixture 190 (2.5mM each), 1μ l of PrimeSTAR GXL DNA Polymerase (1.25U/50μ l) (Takara Bio, Cat# 191 R050A). The PCR extension reaction was used to extend single-stranded long reads containing 192 HPV sequences and the adjacent human genomic sequence. Thermal cycling was performed 193 using SimpliAmp™ Thermal Cycler (Thermo Fisher Scientific Inc., USA) with the following 194 conditions: 98°C for 10 s, annealing temperature (specific to the primer) for 30 s, and 68°C for 6 195 min for 50 cycles. Primer extensions were then size selected using 0.8x ratio of High Prep™ 196 PCR Clean-up Beads (Magbio Genomics, USA) following the manufacturer’s instructions and 197 eluted in 25 μ L of nuclease and protease-free molecular biology grade water (HiMedia Ltd, 198 Mumbai). 199 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint C-Tailing 200 The purified amplification product (100 ng) was initially incubated for 5 min at 95°C in a water 201 bath and chilled immediately on ice for 3 min. The nucleotide tailing reaction was then set up 202 with 5 μ l of 10X TdT buffer, 5 μ l of 2.5 mM CoCl2, 1 μ l of 100 pmol dCTP and 0.5 μ l of 203 terminal transferase (20 units/μ l) (New England Biolabs), and the mixture was incubated for 1 204 hour at 37°C followed by heating for 10 min at 70°C. The reaction mixture was purified with 205 paramagnetic beads and eluted in 10 μ l of distilled water, and 5 μ l was used for the second round 206 PCR. 207 Second round PCR extension 208 The second round PCR included a polyG reverse primer with nanopore-specific adapters for 209 barcoding, 5μ l of C-tailed products of first round of single primer extension, 10μ l 5X 210 PrimeSTAR GXL Buffer, 4μ l dNTP mixture (2.5mM each), 1μ l of PrimeSTAR GXL DNA 211 Polymerase (1.25U/50μ l) (Takara Bio, Cat# R050A). This PCR was performed under the 212 following conditions: 98°C for 30 s, annealing temperature (specific to the primer) for 30 s and 213 68°C for 6-10 min for 40 cycles followed by a 10-min incubation at 68°C. These PCR products 214 were purified using High Prep™ PCR Clean-up System (Magbio Genomics) and eluted in 25μ l. 215 Final PCR enrichment 216 The final round of PCR amplification included primers specific to adapter sequences, 10 μ l 217 purified product from the second round PCR, 10μ l 5x PrimeSTAR GXL Buffer, 4μ l dNTP 218 mixture (2.5mM each), 1μ l of PrimeSTAR GXL DNA Polymerase (1.25U/50μ l) (Takara Bio, 219 Cat# R050A). This PCR was performed with the following conditions: 98°C for 30 s, annealing 220 temperature (specific to the primer) for 30 s, and 68°C for 10 min for 40 cycles followed by a 10 221 min incubation at 68°C. PCR products were purified using High Prep™ PCR Clean-up System 222 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint (Magbio Genomics) and eluted in 25μ l TE buffer. These purified products were processed for 223 PCR barcoding for nanopore sequencing according to the manufacturers’ instructions. The 224 schematic representation of the single primer PCR extension workflow is shown in Figure 1. The 225 list of primers used in this assay is attached in Supplementary Table 1. 226 PCR barcoding 227 PCR barcoding was performed using the barcodes provided in the PCR Barcoding Expansion 1–228 12 kit (EXP-PBC001, Oxford Nanopore Technologies, Oxford, UK). One barcode was used per 229 sample. The barcoding PCR reaction contained 2 μ l PCR Barcode (one of BC01-BC12, at 10 230 μ M), 500ng of purified PCR amplification product, 10μ l of 5X PrimeSTAR GXL Buffer, 4μ l of 231 dNTP mixture (2.5mM each), 1μ l of PrimeSTAR GXL DNA Polymerase (1.25U/50μ l) (Takara 232 Bio, Cat# R050A) and nuclease-free water up to 50 μ L. The cycling conditions used for 233 barcoding PCR consisted of 35 cycles with Initial denaturation 1 cycle of 95 °C for 3 mins, 35 234 cycles of denaturation at 95 °C for 30 seconds, annealing at 62 °C for 30 seconds, extension at 235 65 °C for 6-10 minutes, final extension 1 cycle at 65 °C for 10 minutes and hold at 4 °C. PCR 236 barcoded products were purified using magnetic beads and the concentrations were measured 237 using Qubit® fluorometer 3.0 (Thermo Fisher Scientific Inc.). These products were pooled in 238 equimolar concentrations and 1 μ g of pooled barcoded libraries was diluted in 47 μ L of nuclease-239 free water for nanopore sequencing. 240 241 Nanopore Sequencing 242 The barcoded library was then end-prepped, ligated with adaptors, and cleaned up for 243 sequencing using the ONT sequencing ligation kit SQK-LSK109 kit (Oxford Nanopore 244 Technologies). Qubit® fluorometer 3.0 (Thermo Fisher Scientific Inc.) was used to determine 245 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint the concentration of the generated library. 50 fmol of the prepared library was loaded onto a R9.3 246 flow cell (ONT), after priming the flow cells with 800/i2 µL of priming mix (30/i2 µL Flush Tether 247 to 1.17/i2 mL of Flush Buffer). To prepare the library for loading, 11/i2 µL (50 fmol) of the 248 prepared library was mixed with 34/i2 µL Sequencing Buffer, 25.5/i2 µL pre-mixed loading beads, 249 4.5/i2 µL nuclease-free water. 200/i2 µL of priming mix was added to the priming port again 250 avoiding the introduction of air bubbles. Finally, 75/i2 µL of the sample mix was added to the 251 flow cell SpotON sample port of the R9.3 flow cells (Oxford Nanopore Technologies) on the 252 MinION in a dropwise manner. The samples were sequenced on the MinION Sequencer for 3 253 hours. Fastq files obtained from the nanopore sequencing run were analysed using the 254 nfcore/nanoseq pipeline version 3.1.0 on Nextflow version 22.10.4. The raw fastq files were 255 cleaned using NanoLyse and QC was done using FastQC. The processed reads were mapped to a 256 custom genome consisting of human genome (hg38), HPV16, HPV18 and HPV31 sequences 257 using minimap2 aligner. The chimeric soft clip reads from the output .bam files were extracted 258 for further analysis. Circos plots were generated using the online tool shinyCircos 259 (https://venyao.xyz/shinycircos/) (25). 260 261 Viral Integration Detection Using Nanopore Reads 262 The reads obtained from Nanopore sequencing were aligned to a custom reference genome file 263 using NGMLR- a long read aligner. NGMLR (https://github.com/philres/ngmlr) is designed to 264 quickly and correctly align reads spanning complex structural variations and uses a convex gap 265 cost model to compute precise alignments. It penalizes gap extensions for longer gaps less than 266 for shorter ones thus accounting for large structural variants. The custom reference genome was 267 created by concatenating hg38 and HPV reference genomes (HPV16 & HPV18). The aligned 268 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint bam files (NGMLR output) were then used as input to Sniffles2 269 (https://github.com/fritzsedlazeck/Sniffles), a structural variation caller for long-reads. Sniffles2 270 outputs all types of detected structural variants (deletions, insertions, inversions and breakends). 271 Only the breakends spanning human and viral contigs (with minimum five reads support) were 272 considered as viral integration points. 273 274 Sanger Validation 275 To validate the HPV integration sites in the human genome detected by nanopore sequencing, 276 primers were designed, derived from the human genome at the potential site of integration, and 277 the other against HPV sequences suspected of being near the site of integration within the HPV 278 genome. Both primers were designed 500 bp away from the integration sites detected from the 279 nanopore sequencing results. The PCR reaction mix was prepared in a total volume of 25 μ L 280 containing 12.5 μ L 2X GoTaq Green Master Mix, 0.5 μ L of each forward and reverse primer (10 281 mM), and 10.5 μ L nuclease-free water. 1 μ L (30 ng) genomic DNA solution was used as a 282 template. The PCR conditions were as follows: 5 min at 95◦ C; 35 cycles of 30 s at 94◦ C; 60 s at 283 50–60◦ C for annealing; and 60 s at 72◦ C; followed by 72◦ C for 1 min. The PCR products were 284 run on 1.5 % TAE (Tris base, acetic acid and EDTA) agarose gel, the bands were excised and 285 purified using FavorPrep Gel/PCR purification mini kit (Favorgen Biotech. Corp., Taiwan, 286 Japan). Sequencing of the purified PCR products was performed using a BigDye® Terminator 287 v3.1 Cycle Sequencing Kit (Applied Biosystems, USA). The list of primers used for validation is 288 attached in Supplementary Table 2 289 Functional validation of Impact of HPV integrations on the neighbouring genes: 290 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint We wanted to explore the impact of the HPV integrations on the host gene expression in the 291 context of topologically associating domains (TADs). For shortlisting target genes, cancer-292 related functions were determined from the IntOGen database (26) and existing literature. 293 Promoter-enhancer interaction loop regions were obtained from the GeneHancer database (27). 294 TAD boundaries of HeLa and NHEK cell lines were obtained from existing literature (28). All 295 coordinates were obtained in hg38, or converted from hg19 to hg38 using UCSC LiftOver 296 (https://pubmed.ncbi.nlm.nih.gov/16381938/). Total RNA was extracted from the five cervical 297 cancer frozen tumor tissue and a control fibroblast cell line using TRIzol™ Reagent (Thermo 298 Fisher, Carlsbad, CA, USA) according to the manufacturer’s protocol. 1 μ g of total RNA was 299 converted into cDNA using an iScript™ gDNA Clear cDNA Synthesis Kit (BioRad, Hercules, 300 CA, USA) according to the manufacturer’s instructions. The reaction was incubated at 42°C for 301 30 minutes, followed by 85°C for 5 minutes to inactivate the reverse transcriptase. Primers were 302 designed using primer3 software for qPCR. The list of primers used for qPCR is enlisted in 303 Supplementary Table 3. 304 305 Reverse Transcriptase-Quantitative Polymerase Chain Reaction Analysis of 306 Selected Genes 307 We performed qPCR analysis for the panel of selected genes. To assess whether gene expression 308 changes were specifically associated with HPV integration breakpoints and not due to general 309 HPV infection, each patient was analyzed as a case, with the other four patients serving as 310 controls. This intra-cohort comparative design ensured that differences in gene expression were 311 linked to the integration event and not HPV infection. The 2−ΔΔCt method was used for 312 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint quantification and fold change for the target gene. RT-qPCR analysis was performed using TB 313 Green Premix Ex Taq II (Tli RNase H Plus) (Takara Bio Inc., Tokyo, Japan; RR820A) on 314 QuantStudioTM 5 (Applied Biosystems, Waltham, MA, USA). The PCR cycling conditions were 315 as follows: 95 °C for 1 min followed by 40 cycles of 95 °C for 5 s, annealing temperature 316 (specific to the different genes) for 30s, and 72 °C for 30s. The values were first normalized to 317 the internal reference gene GAPDH, followed by calculating relative expression to healthy 318 control. Statistical significance was inferred using t-test, with a p-value < 0.05 considered 319 statistically significant. 320 321 Sequencing Cost Analysis: 322 The cost per sample was calculated by considering flow cell costs, sequencing kits, wash kits, 323 PCR reagents, and terminal transferase. It also includes all other sample preparation and 324 purification reagents before sequencing. WGS was performed at Medgenome laboratories, India; 325 hence, the price charged per sample for sequencing was considered. 326 327

328 Single primer extensions for detection of HPV integrations in HPV Positive Cell 329 Lines 330 We developed a single primer extension method to capture HPV-human breakpoints by 331 designing primers tiling the genomes of the most commonly occurring high-risk HPV subtypes, 332 HPV16 and HPV18. The single primers allowed us to randomly extend the HPV genome 333 allowing us to capture integrations provided the DNA sequence adjacent to the primer binding 334 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint site is in the vicinity of a HPV-human integration junction. The addition of poly-C tail by the 335 terminal transferase enzyme enables the binding of a reverse primer with poly-G resulting in a 336 double-stranded molecule with nanopore adapters at either end. The reads obtained from the 337 nanopore sequencing were in size ranging from 96 to 4656 base pair (with a median size of 228). 338 On an average, the polyC and polyG tracts observed were 7 to 8 base pairs long. To establish and 339 validate this assay, we used genomic DNA from HPV-positive cervical cancer cell lines, HeLa 340 and SiHa, to map chimeric viral-human breakpoints. Our analysis detected all the previously 341 reported viral-human breakpoints in these cell lines (with >100 reads supporting the integration). 342 We could accurately identify integration sites on chromosome 8 of the HPV18-positive HeLa 343 cell line as reported previously using the DIPS PCR technique Luft et al. (15,29) (Figure 2A, 344 Supplementary Figure 1A). Similarly, using this method, we effectively detected the integration 345 sites on chromosome 13 in the HPV16-positive SiHa cell line as reported by Meissner (30) and 346 Yu et al. (29). To validate our findings, we used the Sanger sequencing approach to ensure the 347 accuracy of detection of viral breakpoints in these cell lines (Figure 2B, Supplementary Figure 348 1B) 349 Characterization of HPV Integrations in tumor samples of cervical cancer patients 350 We used the single primer extension method for analyzing HPV integration sites in five HPV-351 positive cervical cancer patients (details summarized in Table 1). Patient P1 had an integration of 352 the viral genome into intron 4 of the RAD51 paralog B (RAD51B) gene on chromosome 14. We 353 could find only one viral breakpoint within this gene, and the integration resulted from the 354 disruption of L1 gene of the virus. Patient P2, had an integration of HPV genome in a lncRNA 355 gene LINC02046 on chromosome 3. Three patients P3, P5 and P4 had integration sites in 356 intergenic regions of Chromosome 4, and Chromosome 2 respectively. Patients P3 and P5, had 357 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint the exact same integration sites on chromosome 4. WGS was performed for the five patients 358 using Illumina to validate the results of our analysis. Sequencing analysis from WGS 359 demonstrated concordance with nanopore sequencing data (Table 1). We further validated these 360 integration sites by Sanger sequencing (Supplementary Figure 2). The breakpoints obtained for 361 the five patients through Illumina WGS and Nanopore sequencing are mentioned in Table 1 and 362 Supplementary Figure 2. 363 HPV integration has been shown to disrupt host transcriptional activity by upregulating the 364 expression of nearby genes through chromatin remodeling (31). This transcriptional disruption is 365 mostly confined to genes present within the same TAD (Topologically Associated Domain)–366 functional units of 3D genome organization that harbor genes and regulatory regions and 367 spatially confine gene interactions (32). To evaluate the influence of HPV integration on 368 transcriptional activity in the five cervical cancer patients, we shortlisted target genes located 369 within the same TAD as the integration sites (Supplementary Figure 3-6). 370 For patient P1, cancer-associated genes located within the same TAD were selected, including 371 RAD51B, which overlaps with the integration site, and ZFP36L1. In patient P2, the HPV 372 integration occurred near the 5’ TAD boundary. Hence, genes near the 3’ TAD boundary, 373 potentially interacting with the integration site through TAD boundary interactions, were 374 selected. This included CPB1 (within the TAD near its 3’ boundary) and CPA3 (outside the TAD 375 but near its 3’ boundary). For patients P3 and P5, which had the same integration site, genes 376 CXCL8, RASSF6, ANKRD17-DT and MTHFD2L were selected since the integration site fell 377 within their promoter-enhancer integration loops. Additionally, the cancer-associated gene ALB, 378 located within the same TAD, was also selected. No target genes were selected for P4, as the 379 integration site was located outside any TAD, thus devoid of genes or regulatory regions. 380 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint In patient P1, ZFP36L1 was downregulated compared to the control sample and the other four 381 patients, while the expression of RAD51B did not exhibit any significant difference. Genes CPA3 382 and CPB1 were found to be significantly upregulated in patient P2 as compared to the control as 383 well as the other four cervical cancer samples. Patients P3 and P5 shared the same integration 384 breakpoint. Among the cluster of genes selected for P3 and P5, only the relative gene expression 385 CXCL8 was upregulated in both patient P3 and P5 but was normal for the other 3 samples and 386 control. 387 Sequencing Cost 388 The cost per sample for WGS was $598.74 while the final cost of sequencing per sample on 389 MinION using our novel library preparation technique was $27. The final cost was calculated 390 assuming 10 samples per run and did not include labor charges. For calculating the per-run cost 391 for purification, we assumed each sample volume to be 50µL, and in each run, the magnetic bead 392 purification was performed five times. Each flow cell was washed and reused (4 runs per flow 393 cell). When we included the reagent costs for single primer assay along with nanopore 394 sequencing, our cost per-sample was $54.6. Detailed price analysis is included in Table 2. 395

396 Nanopore has emerged as an important next generation sequencing tool since its release in 2014 397 (18). In this study, we developed a novel targeted sequencing approach which combines single 398 primer extension enrichment followed by nanopore sequencing to precisely identify HPV 399 integration sites in HPV-positive cell lines and tumor genomic DNA from cervical cancer patient 400 samples. Whole genome sequencing on the Nanopore sequencing platform has been previously 401 used for studying viral integration due to its ability to generate long reads, enabling 402 comprehensive genome-wide analyses (9). Increasing the depth of sequencing by enriching the 403 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint viral genome using a PCR can help in attaining improved accuracy (21). The findings of our 404 study highlight the effectiveness of our unique targeted nanopore based amplicon sequencing in 405 accurately identifying integration sites within known chromosomal regions of HPV-positive cell 406 lines (HeLa in chr8 and SiHa in chr13). Our results conformed previously identified integration 407 breakpoints in these cell lines (15,16). 408 We further evaluated the utility of our single primer based targeted nanopore sequencing by 409 comparing our results with Illumina Whole Genome Sequencing (WGS). Our analysis of 5 410 cervical cancer patient samples reported concordance with Illumina WGS. In our study, one 411 patient (Patient P1) had an integration within intron 4 of the RAD51B gene located on 412 chromosome 14. RAD51B is an important DNA double-strand break repair protein, and its loss 413 of function is reported in uterine leiomyoma and breast cancer (33). It has also been previously 414 reported to be one of the hotspots for HPV integration which may lead to dysregulation of 415 Homologous Recombination Repair (HRR), causing genomic instability—a hallmark of cancer 416 (34). We also observed that this integration led to the downregulation of a neighboring gene 417 within the same TAD, Zinc finger protein 36 (ZFP36L1), which acts as a tumor suppressor. 418 ZFP36L1 is often downregulated in several patient cohorts of bladder and breast cancers and its 419 reduced expression is associated with worse survival in patients with breast cancer (35). Loss of 420 ZFP36L1 has also been reported to promote epithelial-mesenchymal transition in hepatocellular 421 carcinoma (36). 422 We also found an HPV integration site in a lncRNA gene LINC02046 in Patient P2. The 423 sequencing data for the other three patients revealed integrations predominantly within intergenic 424 regions, which is consistent with patterns reported in previous literature (11,29,37). This 425 integration shared the TAD boundaries with two genes, CPB1 and CPA3, which were found to 426 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint be upregulated in the patient P2. It has been reported that the overexpression of CPB1 in ductal 427 carcinoma in situ (DCIS), which is an early-stage breast cancer, can lead to the progression into 428 invasive breast cancer in patients (38). Also, gene CPA3, currently known as CPA4 (39) is 429 associated with pancreatic cancer progression, and was observed to be overexpressed in 430 pancreatic cancer patients when compared to healthy controls (40). 431 Integration sites in the HPV-positive cell lines HeLa and SiHa were also observed in the 432 intergenic regions of Chr 8 (Chr8q24.21) (11,29) and Chr13 (between the KLF5 and LINC00392 433 genes) (37). As hypothesized by Yang et al., intergenic HPV integrations might serve as a 434 defense mechanism of the virus which keeps the host cell viable as integration into human non-435 exonic regions would probably not cause a significant loss of function of any important gene (9). 436 Interestingly, two patients in our cohort (P3 and P5) had the same intergenic integration site and 437 breakpoint patterns. We observed that this integration break point fell within the promotor 438 enhancer integration loops of genes in that TAD including CXCL8, RASSF6, ANKRD17 and 439 MTHFD2L. CXCL8 was overexpressed in these two patients P3 and P5 as compared to the other 440 three patients and control. None of the other genes, nor ALB, a cancer-associated gene present in 441 the same TAD, showed significant changes in the expression. CXCL8 is a known chemokine 442 which plays major role in the proliferation, invasion, and migration of cancer (41).This gene is 443 also reported to promote angiogenesis in breast cancer (42) and the overexpression of CXCL8 is 444 associated with increased risk of cancer and poorer prognosis in patients with colorectal cancer 445 (43). This integration breakpoint can be further analyzed in a larger cohort of patients for its 446 potential as a biomarker or to be considered as a hotspot. 447 We believe our method has many advantages over both Ilumina and Nanopore WGS. We 448 developed a targeted method for the detection of HPV integration sites which is over 20 times 449 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint more cost-effective than WGS and minimizing sequencing data volume. With our novel 450 enrichment method, we could also enable deeper analysis at the specific region of interest (HPV-451 human Integration breakpoint) which helps to avoid false positive interpretation of sequencing 452 data. Also, this strategic combination not only increased our sensitivity in detecting integration 453 breakpoints but also substantially reduced background noise, enabling us to pinpoint specific 454 integration loci within the genome. 455 The accurate identification and characterization of HPV integration breakpoints holds immense 456 clinical implications, especially in understanding the molecular mechanisms underlying HPV-457 associated carcinogenesis (9). Our study highlights the potential of nanopore sequencing, 458 complemented by the single primer extension enrichment method, as a useful tool for this 459 purpose. Besides the reduced cost and amount of sequencing data, a simpler analysis pipeline 460 makes HPV integration detection faster, more efficient, and easily adoptable, which is crucial for 461 clinical applications. Larger cohorts may be explored in future studies to identify new integration 462 sites, potentially paving the way for the development of prognostic biomarkers or targeted 463 therapeutic approaches for HPV-associated cancers. Our method demonstrates the superiority of 464 targeted nanopore sequencing in identifying HPV integrations compared to Illumina WGS, 465 emphasizing its precision, comprehensive coverage, and compatibility with a novel enrichment 466 method. 467

468 Our study demonstrates the efficacy of a targeted nanopore sequencing approach combined with 469 a novel enrichment method for precise and efficient detection of HPV integrations in genomes of 470 cervical cancer patients. This method has an advantage over the traditional Illumina WGS in 471 terms of cost-effectiveness, and data analysis efficiency. This method will contribute 472 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint significantly to the understanding of the role of HPV integration in cervical carcinogenesis by 473 accurately identifying integration breakpoints. Our novel technique could be easily applied to 474 studying other viral integrations like retroviral, Adenovirus associated and Human Herpes-viral 475 integration events. In addition to mapping viral integrations, this method could also be a crucial 476 and a cost effective solution for detecting gene fusions in several cancers. Larger cohort studies 477 are further required to explore the clinical significance of this method and translate it to 478 diagnostics or companion diagnostic solutions. 479 AUTHORS’ CONTRIBUTIONS 480 PP and RD conceived the project. PP, NM, RS and RD designed the experiments and wrote the 481 manuscript with inputs from all authors. PP and AS2 performed the experiments along with 482 nanopore sequencing. NM, AS1, SM and RS performed nanopore and WGS data analysis. SL 483 and KS assisted in sample characterization and clinical data compilation. DH and MR helped 484 with important inputs to the study design and critical review of the manuscript. RD and RS 485 supervised the study. All authors reviewed and approved the submitted version of the 486 manuscript. 487 ACKNOWLEDGMENTS 488 PP would like to acknowledge the support from Senior Research Fellowship (SRF-Direct) 489 (CSIRAWARD/SRF-DIRECT2024/15079) awarded by Council for Scientific and Industrial 490 Research (CSIR), Ministry of Science & Technology, Government of India. NM would like to 491 acknowledge support from NCBS-TIFR and the Shyama Prasad Mukherjee Fellowship (SPMF) 492 awarded by Council for Scientific and Industrial Research (CSIR), Ministry of Science & 493 Technology, Government of India. RD would like to acknowledge faculty seed money support 494 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint from Manipal Academy of Higher Education, Manipal, Karnataka, India. RD and MR would like 495 to acknowledge the support and fruitful discussions through the Global Cancer Consortium 496 (https://glocacon.org/). 497 FUNDING STATEMENT 498 This study was majorly supported by the Department of Biotechnology, Government of India, 499 Ramalingaswami Fellowship (BT/RLF/Re-entry/21/2018) given to RD. RS would like to 500 acknowledge funding support from the NCBS-TIFR and the DBT/Wellcome Trust India 501 Alliance Fellowship [grant number IA/I/20/1/504928]. 502 CONFLICT OF INTEREST STATEMENT 503 The authors declare no conflict of interest. 504 DATA AVAILABILITY STATEMENT 505 The data that support the findings of this study are available from the corresponding author upon 506 reasonable request. 507 ETHICS STATEMENT 508 Ethical approval for the study was obtained from the Institutional Ethical Committee, Manipal 509 Academy of Higher Education (IEC No: 774/2019). This study was also registered as an 510 observational study with Clinical Trials Registry – India with the number CTRI/2020/01/022862. 511

512 1. Didelot MN, Boulle N, Damay A, Costes V, Segondy M. Comparison of the PapilloCheck® 513 assay with the digene HC2 HPV DNA assay for the detection of 13 high-risk human 514 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint papillomaviruses in cervical and anal scrapes. Journal of Medical Virology. 515 2011;83(8):1377–82. 516 2. Cancer Today [Internet]. [cited 2024 Sep 25]. Available from: https://gco.iarc.who.int/today/ 517 3. Bray F, Laversanne M, Sung H, Ferlay J, Siegel RL, Soerjomataram I, et al. Global cancer 518 statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers 519 in 185 countries. CA: A Cancer Journal for Clinicians. 2024;74(3):229–63. 520 4. Cervical cancer [Internet]. [cited 2024 Sep 25]. Available from: https://www.who.int/health-521 topics/cervical-cancer 522 5. Cervical Cancer Causes, Risk Factors, and Prevention - NCI [Internet]. 2022 [cited 2024 Sep 523 25]. Available from: https://www.cancer.gov/types/cervical/causes-risk-prevention 524 6. Woodman CBJ, Collins SI, Young LS. The natural history of cervical HPV infection: 525 unresolved issues. Nat Rev Cancer. 2007 Jan;7(1):11–22. 526 7. Balasubramaniam SD, Balakrishnan V, Oon CE, Kaur G. Key Molecular Events in Cervical 527 Cancer Development. Medicina (Kaunas). 2019 Jul 17;55(7):384. 528 8. Liu Y, Lu Z, Xu R, Ke Y. Comprehensive mapping of the human papillomavirus (HPV) 529 DNA integration sites in cervical carcinomas by HPV capture technology. Oncotarget. 2015 530 Dec 31;7(5):5852–64. 531 9. Yang W, Liu Y, Dong R, Liu J, Lang J, Yang J, et al. Accurate Detection of HPV Integration 532 Sites in Cervical Cancer Samples Using the Nanopore MinION Sequencer Without Error 533 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint Correction. Frontiers in Genetics [Internet]. 2020 [cited 2023 May 10];11. Available from: 534 https://www.frontiersin.org/articles/10.3389/fgene.2020.00660 535 10. Singh AK, Walavalkar K, Tavernari D, Ciriello G, Notani D, Sabarinathan R. Cis-regulatory 536 effect of HPV integration is constrained by host chromatin architecture in cervical cancers. 537 Mol Oncol. 2024 May;18(5):1189–208. 538 11. Hu Z, Zhu D, Wang W, Li W, Jia W, Zeng X, et al. Genome-wide profiling of HPV 539 integration in cervical cancer identifies clustered genomic hot spots and a potential 540 microhomology-mediated integration mechanism. Nat Genet. 2015 Feb;47(2):158–63. 541 12. Moody CA, Laimins LA. Human papillomavirus oncoproteins: pathways to transformation. 542 Nat Rev Cancer. 2010 Aug;10(8):550–60. 543 13. Zhao JW, Fang F, Guo Y, Zhu TL, Yu YY, Kong FF, et al. HPV16 integration probably 544 contributes to cervical oncogenesis through interrupting tumor suppressor genes and 545 inducing chromosome instability. J Exp Clin Cancer Res. 2016 Nov 25;35(1):180. 546 14. Campitelli M, Jeannot E, Peter M, Lappartient E, Saada S, de la Rochefordière A, et al. 547 Human papillomavirus mutational insertion: specific marker of circulating tumor DNA in 548 cervical cancer patients. PLoS One. 2012;7(8):e43393. 549 15. Luft F, Klaes R, Nees M, Dürst M, Heilmann V, Melsheimer P, et al. Detection of integrated 550 papillomavirus sequences by ligation-mediated PCR (DIPS-PCR) and molecular 551 characterization in cervical cancer cells. International Journal of Cancer. 2001;92(1):9–17. 552 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint 16. Raybould R, Fiander A, Wilkinson GWG, Hibbitts S. HPV integration detection in CaSki 553 and SiHa using detection of integrated papillomavirus sequences and restriction-site PCR. 554 Journal of Virological Methods. 2014 Sep;206:51–4. 555 17. Zhou L, Qiu Q, Zhou Q, Li J, Yu M, Li K, et al. Long-read sequencing unveils high-556 resolution HPV integration and its oncogenic progression in cervical cancer. Nat Commun. 557 2022 May 10;13(1):2563. 558 18. Yang S, Zhao Q, Tang L, Chen Z, Wu Z, Li K, et al. Whole Genome Assembly of Human 559 Papillomavirus by Nanopore Long-Read Sequencing. Frontiers in Genetics [Internet]. 2022 560 [cited 2023 May 15];12. Available from: 561 https://www.frontiersin.org/articles/10.3389/fgene.2021.798608 562 19. Arroyo Mühr LS, Guerendiain D, Cuschieri K, Sundström K. Human Papillomavirus 563 Detection by Whole-Genome Next-Generation Sequencing: Importance of Validation and 564 Quality Assurance Procedures. Viruses. 2021 Jul 8;13(7):1323. 565 20. Chan WS, Chan TL, Au CH, Leung CP, To MY, Ng MK, et al. An economical Nanopore 566 sequencing assay for human papillomavirus (HPV) genotyping. Diagn Pathol. 2020 May 567 6;15(1):45. 568 21. Brancaccio RN, Robitaille A, Dutta S, Rollison DE, Tommasino M, Gheit T. MinION 569 nanopore sequencing and assembly of a complete human papillomavirus genome. Journal of 570 Virological Methods. 2021 Aug 1;294:114180. 571 22. Krueger F. FelixKrueger/TrimGalore [Internet]. 2024 [cited 2024 Sep 25]. Available from: 572 https://github.com/FelixKrueger/TrimGalore 573 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint 23. Nguyen NPD, Deshpande V, Luebeck J, Mischel PS, Bafna V. ViFi: accurate detection of 574 viral integration and mRNA fusion reveals indiscriminate and unregulated transcription in 575 proximal genomic regions in cervical cancer. Nucleic Acids Res. 2018 Apr 20;46(7):3309–576 25. 577 24. PaVE [Internet]. [cited 2024 Sep 25]. Available from: https://pave.niaid.nih.gov/ 578 25. Wang Y, Jia L, Tian G, Dong Y, Zhang X, Zhou Z, et al. shinyCircos/i2 V2.0: Leveraging the 579 creation of Circos plot with enhanced usability and advanced features. Imeta. 2023 May 580 8;2(2):e109. 581 26. Gonzalez-Perez A, Perez-Llamas C, Deu-Pons J, Tamborero D, Schroeder MP, Jene-Sanz A, 582 et al. IntOGen-mutations identifies cancer drivers across tumor types. Nat Methods. 583 2013;10(11):1081–2. 584 27. GeneHancer: genome-wide integration of enhancers and target genes in GeneCards - PMC 585 [Internet]. [cited 2024 Sep 25]. Available from: 586 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5467550/ 587 28. McCole RB, Erceg J, Saylor W, Wu CT. Ultraconserved Elements Occupy Specific Arenas 588 of Three-Dimensional Mammalian Genome Organization. Cell Rep. 2018 Jul 10;24(2):479–589 88. 590 29. Yu L, Majerciak V, Lobanov A, Mirza S, Band V, Liu H, et al. HPV oncogenes expressed 591 from only one of multiple integrated HPV DNA copies drive clonal cell expansion in 592 cervical cancer. mBio. 2024 Apr 16;15(5):e00729-24. 593 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint 30. Meissner JD. Nucleotide sequences and further characterization of human papillomavirus 594 DNA present in the CaSki, SiHa and HeLa cervical carcinoma cell lines. Journal of General 595 Virology. 1999;80(7):1725–33. 596 31. Ojesina AI, Lichtenstein L, Freeman SS, Pedamallu CS, Imaz-Rosshandler I, Pugh TJ, et al. 597 Landscape of genomic alterations in cervical carcinomas. Nature. 2014 Feb 598 20;506(7488):371–5. 599 32. Cis /i2 regulatory effect of HPV integration is constrained by host chromatin architecture in 600 cervical cancers - Singh - 2024 - Molecular Oncology - Wiley Online Library [Internet]. 601 [cited 2024 Sep 25]. Available from: https://febs.onlinelibrary.wiley.com/doi/10.1002/1878-602 0261.13559 603 33. Parfenov M, Pedamallu CS, Gehlenborg N, Freeman SS, Danilova L, Bristow CA, et al. 604 Characterization of HPV and host genome interactions in primary head and neck cancers. 605 Proceedings of the National Academy of Sciences. 2014 Oct 28;111(43):15544–9. 606 34. Li W, Qi Y, Cui X, Huo Q, Zhu L, Zhang A, et al. Characteristic of HPV Integration in the 607 Genome and Transcriptome of Cervical Cancer Tissues. Biomed Res Int. 2018 Jun 608 19;2018:6242173. 609 35. RNA-Binding Protein ZFP36L1 Suppresses Hypoxia and Cell-Cycle Signaling | Cancer 610 Research | American Association for Cancer Research [Internet]. [cited 2024 Sep 25]. 611 Available from: https://aacrjournals.org/cancerres/article/80/2/219/645796/RNA-Binding-612 Protein-ZFP36L1-Suppresses-Hypoxia-and 613 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint 36. Chen J, Patial S. Loss of RNA Binding Protein, ZFP36L1, promotes EMT in hepatocellular 614 cancer cells by regulating EMT-inducing transcription factor ZEB2. The FASEB Journal 615 [Internet]. 2021 [cited 2024 Sep 25];35(S1). Available from: 616 https://onlinelibrary.wiley.com/doi/abs/10.1096/fasebj.2021.35.S1.04347 617 37. Akagi K, Li J, Broutian TR, Padilla-Nash H, Xiao W, Jiang B, et al. Genome-wide analysis 618 of HPV integration in human cancers reveals recurrent, focal genomic instability. Genome 619 Res. 2014 Feb;24(2):185–99. 620 38. Kothari C, Clemenceau A, Ouellette G, Ennour-Idrissi K, Michaud A, Diorio C, et al. Is 621 Carboxypeptidase B1 a Prognostic Marker for Ductal Carcinoma In Situ? Cancers. 2021 Apr 622 6;13(7):1726. 623 39. He K, Zheng J, Zhang T, Lv H, Wang K, Wang Z, et al. Exploring the Potential of CPA4 624 Knockdown as a Prognostic Biomarker in Inhibiting Endometrial Cancer Proliferation 625 [Internet]. 2024 [cited 2024 Sep 25]. Available from: 626 https://www.researchsquare.com/article/rs-4239097/v1 627 40. Sun L, Burnett J, Guo C, Xie Y, Pan J, Yang Z, et al. CPA4 is a promising diagnostic serum 628 biomarker for pancreatic cancer. Am J Cancer Res. 2015 Dec 15;6(1):91–6. 629 41. Liu Q, Li A, Tian Y, Wu JD, Liu Y, Li T, et al. The CXCL8-CXCR1/2 pathways in cancer. 630 Cytokine & Growth Factor Reviews. 2016 Oct 1;31:61–71. 631 42. Azenshtein E, Meshel T, Shina S, Barak N, Keydar I, Ben-Baruch A. The angiogenic factors 632 CXCL8 and VEGF in breast cancer: regulation by an array of pro-malignancy factors. 633 Cancer Lett. 2005 Jan 10;217(1):73–86. 634 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint 43. B ă lăş oiu M, Bă lăş oiu AT, Mogoantă SŞ , Bă rbă lan A, Stepan AE, Ciurea RN, et al. Serum 635 and tumor microenvironment IL-8 values in different stages of colorectal cancer. Rom J 636 Morphol Embryol. 2014;55(2 Suppl):575–8. 637 638 Table 1: Demographic details of patients and comparison of breakpoint obtained by 639 Illumina WGS and Nanopore sequencing 640 Sl No. Patient ID Age Stage of Cancer Histology Integration in Human HPV type & Break- points Illumina WGS Nanopore Sequencing Sanger Sequencing 1 P1 42 IB3 SCC Chr 14: 68500410 (RAD51B gene) HPV 18 (L1) Yes Yes Yes 2 P2 42 IIIC1 SCC Chr 3: 148344900; 148384643 LINC02046 RP11- 455G2.1 HPV 16 (L2,E6 ) Yes Yes Yes .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint 3 P3 42 IIB SCC Chr 4: 73657890; 73718402 (Intergenic) HPV 16 (L2,E1 ) Yes Yes Yes 4 P4 43 IIB SCC Chr2: 145528039; 145678400 (Intergenic) RP11- 707K3.1 HPV 16 (E1, E2) Yes Yes Yes 5 P5 45 IIB SCC Chr 4: 73657890; 73718402 (Intergenic) HPV16 (L2, E1) Yes Yes Yes Footnote: SCC: Squamous Cell Carcinoma 641 642 643 644 645 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint Table 2: Cost analysis for MinION Nanopore sequencing using PCR enrichment method 646 Nanopore Sequencing Cost (in US$) Total Units Cost per run (in US $) PCR 148.9 250 0.6 C-tailing (Terminal Transferase NEB) 213.65 500 0.4 Magnetic beads for purification HighPrep PCR 123.77 125 (40ul per reaction) 5.0 (5 times per run) Minion Flowcell R9.4.1 FLO- MIN106D 1,081.16 4 runs per flow cell 270.3 Ligation Sequencing and adapter ligation (SQK- LSK109 kit) 907.46 6 151.2 Barcoding (PCR Barcoding Expansion 1-12) 454.74 6 75.8 Flowcell Priming (EXP- FLP002) 155.73 6 26.0 Flowcell Wash (EXP- WSH004) 99 6 16.5 Total Cost 545.8 Cost per sample 54.6 647 648 649 650 651 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint Supplementary Table 1: Primers used for Single Primer Enrichment 652 Sl No. Primer Name Primer Sequence (5’-3’) Primer size 1 HPV16E6FOR1 TATGCACAGAGCTGCAAACA 20 2 HPV16E6REV1 GCAAAGTCATATACCTCACGTC 22 3 HPV16E6FOR2 CGGTCGATGTATGTCTTGTT 20 4 HPV 16 E6REV2 CTGGGTTTCTCTACGTGTTC 20 5 HPV16E7REV1 ACAAAGCACACACGTAGACA 20 6 HPV16 E7 F Int TGCAACCAGAGACAACTGAT 20 7 HPV16 E7 R Int TGTCTACGTGTGTGCTTTGT 20 8 HPV16E1REV1 GCGTTGTTTACTGCACAGGA 20 9 HPV16E1REV2 TGCGATTGGTGTATTGCTGC 20 10 HPV16E1REV3 TGATGGAGGTGATTGGAAGCA 21 11 HPV16E1REV5 TGGTACAGATTCTAGGTGGCC 21 12 HPV16E2FOR1 AGTACAGACCTACGTGACCA 20 13 HPV16E2REV1 GGCTAACGTCTTGTAATGTCCA 22 14 HPV16E2FOR2 AAACCCCTGCCACACCACTA 20 15 HPV16E2REV2 TGTCCTGTCCAATGCCATGT 20 16 HPV16E2FOR3 TGCCAACGTTTAAATGTGTG 20 17 HPV16 E2REV3 CGCATGAACTTCCCATACTT 20 18 HPV16L2FOR1 TTGGAACAGGGTCGGGTACA 20 19 HPV 16 L2REV1 CAGATGGTACCGGGGTTGTA 20 20 HPV16L2FOR2 GTCGCACAACACAACAGGTT 20 21 HPV16L1FOR1 CGGATGAATATGTTGCACGC 20 22 HPV16L1REV1 GATCCTTTGCCCCAGTGTTC 20 23 HPV16L1FOR2 CTGGCTTTGGTGCTATGGAC 20 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint 24 HPV16L1REV2 AGGTGCTGGAGGTGTATGTT 20 25 HPV18E6FOR1 ATGGCGCGCTTTGA 14 26 HPV18E6REV1 CTGTAAGTTCCAATACTGTCTTG 23 27 HPV18E7FOR1 TGCATGGACCTAAGGCAAC 19 28 HPV18E7REV1 CACGGACACACAAAGGACAG 20 29 HPV18E7FOR2 GACATTGTATTGCATTTAGAGCC 23 30 HPV18E7REV2 CCATTGTGTGACGTTGTGG 19 31 HPV18E1FOR1 CCACCAAAATTGCGAAGTAGTG 22 32 HPV18E1FOR2 TGTGGACCAGCAAATACAGG 20 33 HPV18E1REV1 ACGGAGGCTATAGACAACG 19 34 HPV18E2REV1 ACGTGGGAAGTACATTTTGGG 21 35 HPV18E2FOR2 ATGTGCAGTACCAGTGACGA 20 37 HPV18L2FOR1 TGGCACGTCTGGGTTTGATA 20 38 HPV18L2FOR2 TGCTTTAACATCCAGGCGTG 20 39 HPV18L2REV1 AGGGTCCATGTCATCTGCAT 20 40 HPV18L1FOR1 ACCTCTGTATGGCCCATTGT 20 41 HPV18L1REV1 AATATGGATTACCAACAGTTAATA A 25 42 HPV18L1REV2 TCACCATCTTCCAAAACTGTGT 23 43 GP5+ TTTGTTACTGTGGTAGATACTAC 23 44 GP6+ GAAAAATAAACTGTAAATCATATT C 25 653 654 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint Supplementary Table 2: List of primers used for validation of breakpoints in cervical cancer 655 samples 656 Sl No. Primer Name Primer Sequence (5’-3’) Primer size 1 P1-F GGACCAAGGAAGTTCAATCAGA 22 2 P1-R TTAGCCCAGTGTTCCCCAAT 20 3 P2(1)-F CCCTGCTTTTATAACCACTCCC 22 4 P2(1)-R AGGCCCCTCACCAATCTGA 19 5 P2(2)-F GCTTTTCTTGCACTCAGTAATGT 23 6 P2(2)-R CGAATGTCTACGTGTGTGCT 20 7 P3/P5(1)-F TCTATGTGAACGGGAACTCTTT 22 8 P3/P5 (1)-R TGGTACCGGGGTTGTAGAAG 20 9 P3/P5 (2)-F CCCTCACACAACTTGCACAA 20 10 P3/P5 (2)-R CAATGGGCCTACGATAATGACA 22 11 P4(1)-F ATATGTTCTCCTGGGGCCTG 20 12 P4(1)-R GGCCACCTAGAATCTGTACCA 21 13 P4(2)-F AAACCCCTGCCACACCACTA 20 14 P4(2)-R GCAAGGCTGACGTAAAGGAT 20 657 658 659 660 661 662 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint Supplementary Table 3: List of primers used for qRTPCR 663 Gene and Primer name Primer Sequence Product Size (bp) ZFP36L1 F TCTGCCACCATCTTCGACTT 151 ZFP36L1 R GGGTGACTGAGTGCCTCC RAD51B F GCACAAAGGTCTGCTGATTTC 182 RAD51B R CCCATGTTGGT GGGTAATGT CXCL8 F CTCTCTTGGCAGCCTTCCT 155 CXCL8 R TGGTCCACTCTCAATCACTCT CPB1 F GTTGGCACTCTTGGTTCTGG 132 CPB1 R GCCAACTCGCGGATTATGTT CPA3 F CCTGTGGGTTTGATTGCTACC 118 CPA3 R TGGCCAAGTCCTTTATGATGTC MTHFD2L F AGAAGCACAGCA CCCTCC 152 MTHFD2L R GATTCCACACCTCGCTGTAT Ankrd17 F TCATCATCACCAGTGGTTTCTTC 181 Ankrd17 R TGTTGATGACTGAAAAGCCAATG RASSF6 F ACGTCTTCTCCAGCAAAGGA 209 RASSF6 R CAGAGCTGCTTCACTCATGG .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint ALB F TCTCTTCTGTCAACCCCACA 127 ALB R CTCTTGTGTGCATCTCGACG GAPDH F CGACCACTTTGTCAAGCTCA 150 GAPDF R GAGGGTCTCTCTCTTCCTCT 664 Figure legends 665 Figure 1: Schematic representation of Single primer extension method 666 (A) Denaturation and hybridization of a single HPV-specific primer to the HPV sequence in the 667 integrated region was followed by single Primer extension by a DNA polymerase; (B) HPV- 668 specific single-stranded extended copies of the target DNA template molecules were then polyC-669 tailed using terminal transferase (TdT); (C) HPV-specific single primer-extended, polyC-tailed 670 ssDNA molecules were then selectively amplified in second round extension with PolyG reverse 671 primer tagged with sequencing adapter; (D) These products were used for library preparation for 672 using adapter-specific primers: complementary to sequencing adapters. (E) Nanopore sequencing 673 was performed and the sequences were analysed for identifying HPV-human integration sites. 674 Created in BioRender. Genetics, M. (2024) BioRender.com/l69i229 675 676 Figure 2: Circos plot illustrating the chimeras between the HPV genome (specific gene) with 677 specific human chromosomes A. Illustrating chimera of HPV 18 with HeLa cell line (in 678 Chromosome 8) and Patient P1 (in Chromosome 14); B. Illustrating chimera of HPV 16 with 679 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint SiHa cell line (in Chromosome 13), Patient P2 (in Chromosome 3), Patient P3 (in Chromosome 680 4), Patient P4 (in Chromosome 2) and Patient P5 (in Chromosome 4). 681 682 Figure 3: Relative expression of genes associated with HPV integration with respect to their 683 TAD domains. A) Relative gene expression of ZFP36L1 and RAD51B. B) Relative gene 684 expression of CPB1 and CPA3. C) Relative gene expression of CXCL8, D) Relative gene 685 expression of ALB, RASSF6, ANKRD17 and MTHFD2L. 686 687 Supplementary figure 1: Depiction of HPV-human junction breakpoints in HPV-positive cell 688 line along with validation with Sanger sequencing. A. HeLa cell line; B. SiHa cell line 689 690 Supplementary figure 2: Depiction of HPV-human junction breakpoints in five HPV-positive 691 cervical cancer patients in the study along with validation with Sanger sequencing. A. Patient P1; 692 B. Patient P2; C. Patient P3; D. Patient P4; E. Patient P5 693 694 Supplementary figure 3: UCSC genome browser views of the HPV integration site in the 695 cervical cancer patient P1. Integration regions as detected by the ViFi algorithm are denoted as 696 red highlighted regions. RAD51B and ZFP36L1 genes were selected for expression analysis. 697 Tracks shown include TAD coordinates from the HeLa and NHEK cell lines, gene transcripts, 698 H3K27Ac marks indicative of cis regulatory elements, and promoter-enhancer interaction loops 699 from the GeneHancer database. 700 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint Supplementary figure 4: UCSC genome browser views of the HPV integration site in the 701 cervical cancer patient P2. Integration regions as detected by the ViFi algorithm are denoted as 702 red highlighted regions. CPB1 and CPA3 genes were selected for expression analysis. Tracks 703 shown include TAD coordinates from the HeLa and NHEK cell lines, gene transcripts, 704 H3K27Ac marks indicative of cis regulatory elements, and promoter-enhancer interaction loops 705 from the GeneHancer database. 706 707 Supplementary figure 5: UCSC genome browser views of the HPV integration site in the 708 cervical cancer patient P3 and P5. CXCL8, RASSF6, ANKRD17-DT, MTHFD2L and ALB 709 genes were selected for expression analysis. Integration regions as detected by the ViFi 710 algorithm are denoted as red highlighted regions. Tracks shown include TAD coordinates from 711 the HeLa and NHEK cell lines, gene transcripts, H3K27Ac marks indicative of cis regulatory 712 elements, and promoter-enhancer interaction loops from the GeneHancer database. 713 714 Supplementary figure 6: UCSC genome browser views of the HPV integration site in the 715 cervical cancer patient P4. No genes were selected for expression analysis. Integration regions as 716 detected by the ViFi algorithm are denoted as red highlighted regions. Tracks shown include 717 TAD coordinates from the HeLa and NHEK cell lines, gene transcripts, H3K27Ac marks 718 indicative of cis regulatory elements, and promoter-enhancer interaction loops from the 719 GeneHancer database. 720 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted October 19, 2024. ; https://doi.org/10.1101/2024.10.17.618842doi: bioRxiv preprint