Transcript Length Research Articles

Generation of Transcript Length Variants and Reprogramming of mRNA Splicing During Atherosclerosis Progression in ApoE-Deficient Mice

Background. Variant 3′UTRs provide mRNAs with different binding sites for miRNAs or RNA-binding proteins (RBPs) allowing the establishment of new regulatory environments. Regulation of 3′UTR length impacts on the control of gene expression by regulating accessibility of miRNAs or RBPs to homologous sequences in mRNAs. Objective. Studying the dynamics of mRNA length variations in atherosclerosis (ATS) progression and reversion in ApoE-deficient mice exposed to a high-fat diet and treated with an αCD40-specific siRNA or with a sequence-scrambled siRNA as control. Methods. We gathered microarray mRNA expression data from the aortas of mice after 2 or 16 weeks of treatments, and used these data in a Bioinformatics analysis. Results. Here, we report the lengthening of the 5′UTR/3′UTRs and the shortening of the CDS in downregulated mRNAs during ATS progression. Furthermore, treatment with the αCD40-specific siRNA resulted in the partial reversion of the 3′UTR lengthening. Exon analysis showed that these length variations were actually due to changes in the number of exons embedded in mRNAs, and the further examination of transcripts co-expressed at weeks 2 and 16 in mice treated with the control siRNA revealed a process of mRNA isoform switching in which transcript variants differed in the patterns of alternative splicing or activated latent/cryptic splice sites. Conclusion. We document length variations in the 5′UTR/3′UTR and CDS of mRNAs downregulated during atherosclerosis progression and suggest a role for mRNA splicing reprogramming and transcript isoform switching in the generation of disease-related mRNA sequence diversity and variability.

De-novo transcriptome of anterior epimorphic regeneration in Perionyx excavatus

Perionyx excavatus, an indigenous earthworm possesses exceptional regeneration capacity. Their anterior regeneration features wound closure, regeneration induction and morphogenesis of damaged organs. This study involved a complete analysis of their transcriptomic dataset, with an emphasis on identifying the genes expressed during regeneration and predicting their implications in the process of regeneration and morphogenesis. Control (first ten segments) and test (5th day blastema) RNA samples of biological replicates were isolated and sequenced on NovaSeq. 6000 using PE150 read length. An average of 98.64% of high-quality data was retained with assembly showing better continuity with the average transcript length with 823 bp and N50 value of 1,858 bp. This is the first report on the comparative transcriptome of P. excavatus during anterior regeneration and this study will shed light on the complexity of annelid regeneration.

Altered mRNA transport and local translation in iNeurons with RNA binding protein knockdown.

Neurons rely on mRNA transport and local translation to facilitate rapid protein synthesis in processes far from the cell body. These processes allow precise spatial and temporal control of translation and are mediated by RNA binding proteins (RBPs), including those known to be associated with neurodegenerative diseases. Here, we use proteomics, transcriptomics, and microscopy to investigate the impact of RBP knockdown on mRNA transport and local translation in iPSC-derived neurons. We find thousands of transcripts enriched in neurites and that many of these transcripts are locally translated, possibly due to the shorter length of transcripts in neurites. Loss of frontotemporal dementia/amyotrophic lateral sclerosis (FTD/ALS)-associated RBPs TDP-43 and hnRNPA1 lead to distinct alterations in the neuritic proteome and transcriptome. TDP-43 knockdown (KD) leads to increased neuritic mRNA and translation. In contrast, hnRNPA1 leads to increased neuritic mRNA, but not translation, and more moderate effects on local mRNA profiles, possibly due to compensation by hnRNPA3. These results highlight the crucial role of FTD/ALS-associated RBPs in mRNA transport and local translation in neurons and the importance of these processes in neuron health and disease.

Long non-coding RNA MAGEA4-AS1 binding to p53 enhances MK2 signaling pathway and promotes the proliferation and metastasis of oral squamous cell carcinoma

Long non-coding RNAs (lncRNAs) regulate the occurrence, development and progression of oral squamous cell carcinoma (OSCC). We elucidated the expression features of MAGEA4-AS1 in patients with OSCC and its activity as an OSCC biomarker. Furthermore, the impact of up-regulation of MAGEA4-AS1 on the cellular behaviors (proliferation, migration and invasion) of OSCC cells and intrinsic signal mechanisms were evaluated. Firstly, we analyzed MAGEA4-AS1 expression data in The Cancer Genome Atlas (TCGA) OSCC using a bioinformatics approach and in 45 pairs of OSCC tissues using qPCR. Then CCK-8, ethynyl deoxyuridine, colony formation, transwell and wound healing assays were conducted to assess changes in the cell proliferation, migration and invasion protential of shMAGEA4-AS1 HSC3 and CAL27 cells. The RNA sequence of MAGEA4-AS1 was identified using the rapid amplification of cDNA ends (RACE) assay. And whole-transcriptome sequencing was used to identify MAGEA4-AS1 affected genes. Additionally, dual-luciferase reporter system, RNA-binding protein immunoprecipitation (RIP), and rescue experiments were performed to clarify the role of the MAGEA4-AS1-p53-MK2 signaling pathway. As results, we found MAGEA4-AS1 was up-regulated in OSCC tissues. We identified a 418 nucleotides length of the MAGEA4-AS1 transcript and it primarily located in the cell nucleus. MAGEA4-AS1 stable knockdown weakened the proliferation, migration and invasion abilities of OSCC cells. Mechanistically, p53 protein was capable to activate MK2 gene transcription. RIP assay revealed an interaction between p53 and MAGEA4-AS1. MK2 up-regulation in MAGEA4-AS1 down-regulated OSCC cells restored MK2 and epithelial-to-mesenchymal transition related proteins’ expression levels. In conclusion, MAGEA4-AS1-p53 complexes bind to MK2 promoter, enhancing the transcription of MK2 and activating the downstream signaling pathways, consequently promoting the proliferation and metastasis of OSCC cells. MAGEA4-AS1 may serve as a diagnostic marker and therapeutic target for OSCC patients.

Eukaryotic initiation factor 4B is a multi-functional RNA binding protein that regulates histone mRNAs.

RNA binding proteins drive proliferation and tumorigenesis by regulating the translation and stability of specific subsets of messenger RNAs (mRNAs). We have investigated the role of eukaryotic initiation factor 4B (eIF4B) in this process and identify 10-fold more RNA binding sites for eIF4B in tumour cells from patients with diffuse large B-cell lymphoma compared to control B cells and, using individual-nucleotide resolution UV cross-linking and immunoprecipitation, find that eIF4B binds the entire length of mRNA transcripts. eIF4B stimulates the helicase activity of eIF4A, thereby promoting the unwinding of RNA structure within the 5' untranslated regions of mRNAs. We have found that, in addition to its well-documented role in mRNA translation, eIF4B additionally interacts with proteins associated with RNA turnover, including UPF1 (up-frameshift protein 1), which plays a key role in histone mRNA degradation at the end of S phase. Consistent with these data, we locate an eIF4B binding site upstream of the stem-loop structure in histone mRNAs and show that decreased eIF4B expression alters histone mRNA turnover and delays cell cycle progression through S phase. Collectively, these data provide insight into how eIF4B promotes tumorigenesis.

Characterization and functional insights of the novel RC-type plasmid pAnox1 from Anoxybacillus gonensis 05S15

The plasmid pAnox1, isolated from Anoxybacillus gonensis 05S15, was sequenced and characterized as a circular, double-stranded DNA molecule of 1592 base pairs with a GC content of 40.01 %. Despite its cryptic nature and small genome, bioinformatic analyses identified conserved motifs associated with replication-related proteins, though BLAST searches revealed no significant homology with other plasmids. The plasmid genome contains five putative Open Reading Frames (ORFs), four palindromic sequences, and two direct repeats on both strands, suggesting regulatory roles. Electron microscopy and Southern hybridization studies confirmed that pAnox1 follows a Rolling Circle (RC) replication mode. The study further demonstrated that the plasmid encodes three distinct transcripts: ORF-1 and ORF-3 are oriented in the same direction, while ORF-5 is on the opposite strand. RACE and LACE analyses revealed transcript lengths of 903 bp for ORF1, 499 bp for ORF3, and 211 bp for ORF5. Quantitative real-time PCR estimated the relative copy number of pAnox1 at 127 ± 2 copies per chromosomal equivalent. This novel RC-type plasmid in the Anoxybacillus genome holds promise as a cloning and expression vector for biotechnological applications and in vivo protein engineering.

Selection on genome-wide gene expression plasticity of rice in wet and dry field environments.

Gene expression can be highly plastic in response to environmental variation. However, we know little about how expression plasticity is shaped by natural selection and evolves in wild and domesticated species. We used genotypic selection analysis to characterize selection on drought-induced plasticity of over 7,500 leaf transcripts of 118 rice accessions (genotypes) from different environmental conditions grown in a field experiment. Gene expression plasticity was neutral for most gradually plastic transcripts, but transcripts with discrete patterns of expression showed stronger selection on expression plasticity. Whether plasticity was adaptive and co-gradient or maladaptive and counter-gradient varied among varietal groups. No transcripts that experienced selection for plasticity across environments showed selection against plasticity within environments, indicating a lack of evidence for costs of adaptive plasticity that may constrain its evolution. Selection on expression plasticity was influenced by degree of plasticity, transcript length and gene body methylation. We observed positive selection on plasticity of co-expression modules containing transcripts involved in photosynthesis, translation and responsiveness to abiotic stress. Taken together, these results indicate that patterns of selection on expression plasticity were context-dependent and likely associated with environmental conditions of varietal groups, but that the evolution of adaptive plasticity would likely not be constrained by opposing patterns of selection on plasticity within compared to across environments. These results offer a genome-wide view of patterns of selection and ecological constraints on gene expression plasticity and provide insights into the interplay between plastic and evolutionary responses to drought at the molecular level.

Long-Read MDM4 Sequencing Reveals Aberrant Isoform Landscape in Metastatic Melanomas.

MDM4 is upregulated in the majority of melanoma cases and has been described as a "key therapeutic target in cutaneous melanoma". Numerous isoforms of MDM4 exist, with few studies examining their specific expression in human tissues. The changes in splicing of MDM4 during human melanomagenesis are critical to p53 activity and represent potential therapeutic targets. Compounding this, studies relying on short reads lose "connectivity" data, so full transcripts are frequently only inferred from the presence of splice junction reads. To address this problem, long-read nanopore sequencing was utilized to read the entire length of transcripts. Here, MDM4 transcripts, both alternative and canonical, are characterized in a pilot cohort of human melanoma specimens. RT-PCR was first used to identify the presence of novel splice junctions in these specimens. RT-qPCR then quantified the expression of major MDM4 isoforms observed during sequencing. The current study both identifies and quantifies MDM4 isoforms present in melanoma tumor samples. In the current study, we observed high expression levels of MDM4-S, MDM4-FL, MDM4-A, and the previously undescribed Ensembl transcript MDM4-209. A novel transcript lacking both exons 6 and 9 is observed and named MDM4-A/S for its resemblance to both MDM4-A and MDM4-S isoforms.

Navigating the labyrinth of long non-coding RNAs in colorectal cancer: From chemoresistance to autophagy

Long non-coding RNAs (lncRNAs), with transcript lengths exceeding 200 nucleotides and little or no protein-coding capacity, have been found to impact colorectal cancer (CRC) through various biological processes. LncRNA expression can regulate autophagy, which plays dual roles in the initiation and progression of cancers, including CRC. Abnormal expression of lncRNAs is associated with the emergence of chemoresistance. Moreover, it has been confirmed that targeting autophagy through lncRNA regulation could be a viable approach for combating chemoresistance. Two recent studies titled “Human β-defensin-1 affects the mammalian target of rapamycin pathway and autophagy in colon cancer cells through long non-coding RNA TCONS_00014506” and “Upregulated lncRNA PRNT promotes progression and oxaliplatin resistance of colorectal cancer cells by regulating HIPK2 transcription” revealed novel insights into lncRNAs associated with autophagy and oxaliplatin resistance in CRC, respectively. In this editorial, we particularly focus on the regulatory role of lncRNAs in CRC-related autophagy and chemoresistance since the regulation of chemotherapeutic sensitivity by intervening with the lncRNAs involved in the autophagy process has become a promising new approach for cancer treatment.

High throughput long-read sequencing of circulating lymphocytes of the evolutionarily distant sea lamprey reveals diversity and common elements of the variable lymphocyte receptor B (VLRB) repertoire.

The leucine-rich repeat-based variable lymphocyte receptor B (VLRB) antibody system of jawless vertebrates is capable of generating an antibody repertoire equal to or exceeding the diversity of antibody repertoires of jawed vertebrates. Unlike immunoglobulin-based immune repertoires, the VLRB repertoire diversity is characterized by variable lengths of VLRB encoding transcripts, rendering conventional immunoreceptor repertoire sequencing approaches unsuitable for VLRB repertoire sequencing. Here we demonstrate that long-read single-molecule real-time (SMRT) sequencing (PacBio) approaches permit the efficient large-scale assessment of the VLRB repertoire. We present a computational pipeline for sequence data processing and provide the first repertoire-based analysis of VLRB protein characteristics including properties of its subunits and regions of diversity within each structural leucine-rich repeat subunit. Our study provides a template to explore changes in the VLRB repertoire during immune responses and to establish large scale VLRB repertoire databases for computational approaches aimed at isolating monoclonal VLRB reagents for biomedical research and clinical applications.

Evolution of retrocopies in the context of HUSH silencing

Retrotransposition is one of the main factors responsible for gene duplication and thus genome evolution. However, the sequences that undergo this process are not only an excellent source of biological diversity, but in certain cases also pose a threat to the integrity of the DNA. One of the mechanisms that protects against the incorporation of mobile elements is the HUSH complex, which is responsible for silencing long, intronless, transcriptionally active transposed sequences that are rich in adenine on the sense strand. In this study, broad sets of human and porcine retrocopies were analysed with respect to the above factors, taking into account evolution of these molecules. Analysis of expression pattern, genomic structure, transcript length, and nucleotide substitution frequency showed the strong relationship between the expression level and exon length as well as the protective nature of introns. The results of the studies also showed that there is no direct correlation between the expression level and adenine content. However, protein-coding retrocopies, which have a lower adenine content, have a significantly higher expression level than the adenine-rich non-coding but expressed retrocopies. Therefore, although the mechanism of HUSH silencing may be an important part of the regulation of retrocopy expression, it is one component of a more complex molecular network that remains to be elucidated.

RNA-Independent Regulatory Functions of lncRNA in Complex Disease.

During the metagenomics era, high-throughput sequencing efforts both in mice and humans indicate that non-coding RNAs (ncRNAs) constitute a significant fraction of the transcribed genome. During the past decades, the regulatory role of these non-coding transcripts along with their interactions with other molecules have been extensively characterized. However, the study of long non-coding RNAs (lncRNAs), an ncRNA regulatory class with transcript lengths that exceed 200 nucleotides, revealed that certain non-coding transcripts are transcriptional "by-products", while their loci exert their downstream regulatory functions through RNA-independent mechanisms. Such mechanisms include, but are not limited to, chromatin interactions and complex promoter-enhancer competition schemes that involve the underlying ncRNA locus with or without its nascent transcription, mediating significant or even exclusive roles in the regulation of downstream target genes in mammals. Interestingly, such RNA-independent mechanisms often drive pathological manifestations, including oncogenesis. In this review, we summarize selective examples of lncRNAs that regulate target genes independently of their produced transcripts.

Molecular mechanisms uncovers differentially expressed genes in male and female adults of Batocera horsfieldi (Coleoptera: Cerambycidae)

Batocera horsfieldi (Hope) (Coleoptera: Cerambycidae) is a significant wood-boring pest in China, primarily affecting timber forests and economic plantations. However, the genomic resources of B. horsfieldi remain limited, constraining the ability to discover molecular and genetic mechanisms for pest control. This study used RNA-seq to correct PacBio Iso-Seq data, thoroughly analyzing the differential genes of male and female adult B. horsfieldi to gain a more comprehensive understanding of the molecular characteristics of this pest. The sequencing results generated 41.16 GB of raw data and 38.26 GB of clean data, yielding 42,005 full-length transcripts. The average transcript length was 2,609 bp, with an N50 length of 2,858 bp and an N90 length of 1,714 bp. Expression analysis revealed 208 differentially expressed genes (DEGs). Enrichment analysis of these DEGs using Gene Ontology (GO) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) indicated their involvement in energy metabolism, biosynthesis, and cell signal transduction pathways. These pathways collectively maintain metabolic balance, signal transduction networks, and adaptability to external environments within the organism. Quantitative real-time PCR (qRT-PCR) was used for further validation of gene expression levels, and the results were consistent with RNA-seq. The transcriptome data obtained in this study provide additional valuable information for unraveling the molecular mechanisms of B. horsfieldi and offer potential reference points for pest control efforts.

Co-transcriptional splicing facilitates transcription of gigantic genes.

Although introns are typically tens to thousands of nucleotides, there are notable exceptions. In flies as well as humans, a small number of genes contain introns that are more than 1000 times larger than typical introns, exceeding hundreds of kilobases (kb) to megabases (Mb). It remains unknown why gigantic introns exist and how cells overcome the challenges associated with their transcription and RNA processing. The Drosophila Y chromosome contains some of the largest genes identified to date: multiple genes exceed 4Mb, with introns accounting for over 99% of the gene span. Here we demonstrate that co-transcriptional splicing of these gigantic Y-linked genes is important to ensure successful transcription: perturbation of splicing led to the attenuation of transcription, leading to a failure to produce mature mRNA. Cytologically, defective splicing of the Y-linked gigantic genes resulted in disorganization of transcripts within the nucleus suggestive of entanglement of transcripts, likely resulting from unspliced long RNAs. We propose that co-transcriptional splicing maintains the length of nascent transcripts of gigantic genes under a critical threshold, preventing their entanglement and ensuring proper gene expression. Our study reveals a novel biological significance of co-transcriptional splicing.

Transcriptomic profiling reveals salt-responsive long non-coding RNAs and their putative target genes for improving salt tolerance in upland cotton (Gossypium hirsutum)

Salt stress continues to exert notable impacts on the growth and development of cotton plants. In various biological processes of plants and in response to abiotic stress, long noncoding RNAs (lncRNAs) play a crucial regulatory role. In this study, we conducted transcriptome sequencing of cotton plants grown under salt stress and normal water conditions. A total of 519 lncRNAs and 75,376 messenger RNAs (mRNAs) were identified. Compared with mRNAs, lncRNAs have shorter transcript lengths and open reading frames, fewer exons, and lower expression levels. By comparing and analyzing the differentially expressed lncRNAs (DELs) and differentially expressed coding genes (DEGs) after salt treatment, we identified a total of 60 DELs and 217 DEGs. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses revealed that these target genes are enriched primarily in intracellular organelles and participate in metabolic regulation pathways. Ultimately, lncRNA XLOC_043671, along with its target gene Gohir.D02G125200, and the lncRNA XLOC_034831, along with its target gene Gohir.A12G114800 was identified as a participant in salt tolerance regulation in cotton. RNA-seq and RTqPCR revealed that both these lncRNAs and their target genes exhibited significantly enhanced expression under salt stress. Subsequently, utilizing virus-induced gene silencing (VIGS) technology, we successfully silenced the two lncRNAs in cotton plants, which resulted in reduced expression levels of the corresponding target genes, rendering the plants less salt tolerant than did the control plants. Similarly, we successfully silenced these two target genes in cotton seedlings, and the silenced plants demonstrated a pronounced salt-sensitive phenotype. In summary, our transcriptome analysis revealed a significant number of differentially expressed lncRNAs and their corresponding coding genes. These findings lay the foundation for further investigating the mechanisms of lncRNAs and their target genes in the salt stress response of cotton.

Regulatory roles of long non-coding RNAs in short-term heat stress in adult worker bees

Long non-coding RNAs (lncRNAs) are crucial modulators of post-transcriptional gene expression regulation, cell fate determination, and disease development. However, lncRNA functions during short-term heat stress in adult worker bees are poorly understood. Here, we performed deep sequencing and bioinformatic analyses of honeybee lncRNAs. RNA interference was performed by using siRNA targeting the most highly expressed lncRNA. The silencing effect on lncRNA and the relative expression levels of seven heat shock protein (HSP) genes, were subsequently examined. Overall, 7,842 lncRNAs and 115 differentially expressed lncRNAs (DELs) were identified in adult worker bees following heat stress exposure. Structural analysis revealed that the overall expression abundance, length of transcripts, exon number, and open reading frames of lncRNAs were lower than those of mRNAs. GO analysis revealed that the target genes were mainly involved in “metabolism,” “protein folding,” “response to stress,” and “signal transduction” pathways. KEGG analysis indicated that the “protein processing in endoplasmic reticulum” and “longevity regulating pathway-multiple species” pathways were most enriched. Quantitative real-time polymerase chain reaction (qRT-PCR) detection of the selected DELs confirmed the reliability of the sequencing data. Moreover, the siRNA experiment indicated that feeding siRNA yielded a silencing efficiency of 77.51% for lncRNA MSTRG.9645.5. Upon silencing this lncRNA, the expression levels of three HSP genes were significantly downregulated (p < 0.05), whereas those of three other HSP genes were significantly upregulated (p < 0.05). Our results provide a new perspective for understanding the regulatory mechanisms of lncRNAs in adult worker bees under short-term heat stress.

Quantifying 3′UTR length from scRNA-seq data reveals changes independent of gene expression

Although more than half of all genes generate transcripts that differ in 3′UTR length, current analysis pipelines only quantify the amount but not the length of mRNA transcripts. 3′UTR length is determined by 3′ end cleavage sites (CS). We map CS in more than 200 primary human and mouse cell types and increase CS annotations relative to the GENCODE database by 40%. Approximately half of all CS are used in few cell types, revealing that most genes only have one or two major 3′ ends. We incorporate the CS annotations into a computational pipeline, called scUTRquant, for rapid, accurate, and simultaneous quantification of gene and 3′UTR isoform expression from single-cell RNA sequencing (scRNA-seq) data. When applying scUTRquant to data from 474 cell types and 2134 perturbations, we discover extensive 3′UTR length changes across cell types that are as widespread and coordinately regulated as gene expression changes but affect mostly different genes. Our data indicate that mRNA abundance and mRNA length are two largely independent axes of gene regulation that together determine the amount and spatial organization of protein synthesis.

Automated translation accurately translates recorded pediatric neurosurgery clinic conversations between Spanish and English.

The purpose of this study is to analyze an automated voice to text translation device by reporting the translation accuracy for recorded pediatric neurosurgery clinic conversations, classifying errors in translation according to their impact on overall understanding, and comparing the incidence of these errors in English to Spanish vs. Spanish to English conversations. English and Spanish speaking patients at a single academic health system's outpatient pediatric neurosurgery clinic had their conversations recorded. These recordings were played back to a Google Pixel handheld smartphone with Live Translate voice to text translation software. A certified medical interpreter evaluated recordings for incidence of minor errors, errors impacting understanding, and catastrophic errors affecting patient-provider relationship or care. Two proportion t-testing was used to compare these outcomes. 50 patient visits were recorded: 40 English recordings translated to Spanish and 10 Spanish recordings translated to English. The mean transcript length was 4244 ± 992 words. The overall accuracy was 98.2% ± 0.5%. On average, 46 words were missed in translation (1.09% error rate), 31 understanding-altering translation errors (0.73% error rate), and 0 catastrophic errors were made. There was no significant difference in English to Spanish or vice versa. Voice to text translation devices using automatic speech recognition accurately translate recorded clinic conversations between Spanish and English with high accuracy and low incidence of errors impacting medical care or understanding. Further study should investigate additional languages, assess patient preferences and potential concerns with respect to device use, and compare these devices directly to medical interpreters in live clinic settings.

Using ChatGPT-4 to Create Structured Medical Notes From Audio Recordings of Physician-Patient Encounters: Comparative Study.

Medical documentation plays a crucial role in clinical practice, facilitating accurate patient management and communication among health care professionals. However, inaccuracies in medical notes can lead to miscommunication and diagnostic errors. Additionally, the demands of documentation contribute to physician burnout. Although intermediaries like medical scribes and speech recognition software have been used to ease this burden, they have limitations in terms of accuracy and addressing provider-specific metrics. The integration of ambient artificial intelligence (AI)-powered solutions offers a promising way to improve documentation while fitting seamlessly into existing workflows. This study aims to assess the accuracy and quality of Subjective, Objective, Assessment, and Plan (SOAP) notes generated by ChatGPT-4, an AI model, using established transcripts of History and Physical Examination as the gold standard. We seek to identify potential errors and evaluate the model's performance across different categories. We conducted simulated patient-provider encounters representing various ambulatory specialties and transcribed the audio files. Key reportable elements were identified, and ChatGPT-4 was used to generate SOAP notes based on these transcripts. Three versions of each note were created and compared to the gold standard via chart review; errors generated from the comparison were categorized as omissions, incorrect information, or additions. We compared the accuracy of data elements across versions, transcript length, and data categories. Additionally, we assessed note quality using the Physician Documentation Quality Instrument (PDQI) scoring system. Although ChatGPT-4 consistently generated SOAP-style notes, there were, on average, 23.6 errors per clinical case, with errors of omission (86%) being the most common, followed by addition errors (10.5%) and inclusion of incorrect facts (3.2%). There was significant variance between replicates of the same case, with only 52.9% of data elements reported correctly across all 3 replicates. The accuracy of data elements varied across cases, with the highest accuracy observed in the "Objective" section. Consequently, the measure of note quality, assessed by PDQI, demonstrated intra- and intercase variance. Finally, the accuracy of ChatGPT-4 was inversely correlated to both the transcript length (P=.05) and the number of scorable data elements (P=.05). Our study reveals substantial variability in errors, accuracy, and note quality generated by ChatGPT-4. Errors were not limited to specific sections, and the inconsistency in error types across replicates complicated predictability. Transcript length and data complexity were inversely correlated with note accuracy, raising concerns about the model's effectiveness in handling complex medical cases. The quality and reliability of clinical notes produced by ChatGPT-4 do not meet the standards required for clinical use. Although AI holds promise in health care, caution should be exercised before widespread adoption. Further research is needed to address accuracy, variability, and potential errors. ChatGPT-4, while valuable in various applications, should not be considered a safe alternative to human-generated clinical documentation at this time.

Abstract 2323: Deciphering RNA degradation: Insights from a comparative analysis of paired fresh frozen/FFPE total RNA-seq

Abstract Background: Fresh frozen (FF) and formalin-fixed paraffin-embedded (FFPE) samples are primary resources for archival tissues in cancer studies. Despite the advantages of easy storage and cost-effectiveness, FFPE samples have the disadvantage of inevitable chemical-induced RNA degradation. Although, in mRNA-seq data, the 3' bias due to the characteristics of the mRNA-seq platform allows the measure of RNA degradation levels, for total RNA-seq and FFPE samples, there is still no clear measure for RNA degradation. Methods: Our study, utilizing The Cancer Genome Atlas (TCGA) pilot data comprised of 26 paired samples of fresh frozen mRNA-seq [FFM], fresh frozen total RNA-seq [FFT], and FFPE total RNA-seq [PET], investigated RNA degradation patterns in FFPE samples. Upon investigating the read coverage of transcripts in FFPE data, we observed increased noise parameters in selected samples compared to others, suggesting a method for assessment of RNA degradation in FFPE. To measure these noise patterns, we developed a method called windowCV (wCV) utilizing coefficient of variance (CV) along the transcript length. Also, based on reports that membrane-surrounded RNA such as lncRNA and mitochondrial RNA (mtRNA) might be less affected by formalin fixation, we developed the simple measure dividing expression value of other proteins with certain lncRNA or mtRNA inferring the degree of RNA degradation. Results: We first compared the shape of transcript read coverage from 3 different RNA-seq data and observed 3’bias pattern in FFM which was not observed in FFT and PET. For FFT and PET, we applied wCV and executed hierarchical clustering. We observed that a subset of samples and genes were clustered by high wCV value suggesting notable RNA degradation. To support the hypothesis of RNA degradation, we utilized Trimmed Mean of M values (TMM) expression ratio of paired FFT and PET to determine which samples and genes were associated with underestimation of expression in PET. Interestingly, we observed that PET/FFT TMM ratio showed negative correlation with wCV. Additionally, when we investigated hierarchical clustering result of PET/FFT TMM ratio, we found that lncRNA and mtRNA were less affected by FFPE. We then calculated expression ratio between membrane-surrounded RNA and a set of protein coding genes largely affected by FFPE. As a result, we found that this ratio positively correlated with paired PET/FFT TMM ratio. We conclude that this simple measurement can provide information of FFPE induced underestimated expression with limited transcriptome data as well as RNA degradation. Conclusion: Our method can provide gene-level CV while avoiding false positive variance signal originating from transcript model and simple way to detect the degree of RNA degradation across FFPE samples. Citation Format: Wonyoung Choi, Miyeon Yeon, Hyo Young Choi, David Neil Hayes. Deciphering RNA degradation: Insights from a comparative analysis of paired fresh frozen/FFPE total RNA-seq [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2323.