AI-driven framework enables precise prediction of RNA splicing and isoform usage

Phys · 🇨🇳 public China · May 19, 2026 · 632 words · By Chinese Academy

Cancer Research Genomics and RNA Splicing Artificial Intelligence in Medicine

headphones Listen to the eFinder podcast briefing

Generate a natural audio summary of this story

Daily briefing

What to know about Cancer Research

Researchers from the China National Center for Bioinformation have developed an AI-driven framework called HELIX and its single-cell variant scHELIX to predict RNA splicing and isoform usage. The study, published in Nature Computational Science, demonstrates the model's ability to outperform existing methods and identify splicing dysregulation in colorectal cancer.

Propaganda risk 10%

Claims checked 9

Techniques found 1

Topics 3

Coverage spectrum

Coverage gap: Low Left coverage

Left0%

Center100%

Right0%

2 sources compared across this story cluster. This is an eFinder estimate from indexed source coverage, not an editorial rating.

What happened

AI-driven framework enables precise prediction of RNA splicing and isoform usage Gaby Clark Scientific Editor Robert Egan Associate Editor RNA is the means of translating the genetic code embedded in DNA into proteins, which serve as enzymes, transporters,…

Why it matters

Yet one gene is not limited to producing one RNA variant.

Common ground

The process of RNA splicing—in which different coding RNA segments (exons) are joined together after noncoding regions (introns) are removed—allows for the generation of a large array of RNA transcript isoforms with distinct sequences, and consequently,…

Perspective signals

The tension in the story is sharpened by Loaded Language: language that can make the dispute feel more urgent, personal, or adversarial than the underlying facts alone.

Follow-up questions

What new context would change how readers understand this Cancer Research story?
What evidence would most clearly confirm or weaken the claim that researchers from the China National Center for Bioinformation, a research center affiliated to the Chinese Academy of Sciences, led by Professor Gao Yuan, have developed an AI-driven framework that enables highly accurate prediction of RNA splicing and isoform usage by integrating genomic sequence features with tissue-specific RBP expression profiles?
How does this story connect Cancer Research with Genomics and RNA Splicing over the next few days?

open_in_new Read the original article: https://phys.org/news/2026-05-ai-driven-framework-enables-precise.html

analyticsAnalysis

10%

Propaganda Score

confidence: 95%

Low risk. This article shows minimal use of propaganda techniques.

psychologyPropaganda Techniques Detected

eFinder identified 1 propaganda technique in this article. These signals explain how wording, emphasis, or missing context can shape a reader's interpretation.

warning

Loaded Language 70% confidence

Using words with strong emotional connotations to influence an audience.

Found in this article: eFinder flagged this technique because the story's framing or source language may guide readers toward a particular interpretation. Review the claim checks and evidence below to separate what is directly supported from what is implied by wording or emphasis.

Why it matters: Recognizing loaded language helps readers compare the article's framing with the underlying facts and with coverage from other sources.

fact_checkClaims Checked

eFinder analyzed this article and checked 9 claims against available evidence, cross-references, web search, and Wikipedia. Here is what the fact-checking layer found.

info Single Source 7

help Insufficient Evidence 1

verified Verified By Reference 1

info

Claim 1: “researchers from the China National Center for Bioinformation, a research center affiliated to the Chinese Academy of Sciences, led by Professor Gao Yuan, have developed an AI-driven framework that enables highly accurate prediction of RNA splicing and isoform usage by integrating genomic sequence features with tissue-specific RBP expression profiles.”

SINGLE SOURCE

While a web search result mentions an 'AI-driven framework' for RNA splicing and RBP expression profiles, the specific details regarding Professor Gao Yuan and the China National Center for Bioinformation are not corroborated by a second independent source in the provided evidence.

menu_book

wikipedia NEUTRAL — The Han Chinese, alternatively Han people, or Chinese people, are an East Asian ethnic group native to Greater China. With a global population of over 1.4 billion, the Han Chinese are the world's larg…
https://en.wikipedia.org/wiki/Han_Chinese

menu_book

wikipedia NEUTRAL — The Mongol conquest of China was a series of major military efforts by the Mongol Empire to conquer various empires ruling over China for 74 years (1205–1279). It spanned over seventy years in the 13t…
https://en.wikipedia.org/wiki/Mongol_conquest_of_China

menu_book

wikipedia NEUTRAL — Gao Shangquan (September 10, 1929 – June 27, 2021) was a Chinese economist.
https://en.wikipedia.org/wiki/Gao_Shangquan

+ 3 more evidence sources

info

Claim 2: “Results show that HELIX substantially outperforms existing mainstream methods in both splicing strength prediction and overall isoform usage prediction.”

SINGLE SOURCE

One web search result explicitly states that HELIX substantially outperforms existing mainstream methods in splicing strength and isoform usage prediction.

travel_explore

web search NEUTRAL — Results show that HELIX substantially outperforms existing mainstream methods in both splicing strength prediction and overall isoform usage prediction.HELIX: a scalable model for predicting context-d…
https://www.brightsurf.com/news/8OMP43N1/ai-driven-framework…

travel_explore

web search NEUTRAL — Description: This video reviews the process of splicing, and discusses in silico splice site predictors, demonstrating their use in the interpretation of spl...
https://www.youtube.com/watch?v=7uI5KWpeBRU

travel_explore

web search NEUTRAL — Biochemical and phosphoproteomic analysis of the helix-loop-helix protein E47.
https://www.tandfonline.com/journals/tmcb20

info

Claim 3: “It first integrates DNA sequence information with the expression profiles of 1,499 RBPs and then employs long short-term memory (LSTM) networks”

SINGLE SOURCE

One web search result confirms the framework integrates genomic sequence features with RBP expression profiles, but the specific number '1,499 RBPs' and the use of LSTM networks for this specific integration are not corroborated by another independent source.

menu_book

wikipedia NEUTRAL — Deoxyribonucleic acid (; DNA) is a polymer composed of two polynucleotide chains that coil around each other to form a double helix. The polymer carries genetic instructions for the development, funct…
https://en.wikipedia.org/wiki/DNA

menu_book

wikipedia NEUTRAL — DNA replication is the process by which a cell makes exact copies of its DNA. This process occurs in all organisms and is essential to biological inheritance, cell division, and repair of damaged tiss…
https://en.wikipedia.org/wiki/DNA_replication

menu_book

wikipedia NEUTRAL — In the fields of geometry and biochemistry, a triple helix (pl.: triple helices) is a set of three congruent geometrical helices with the same axis, differing by a translation along the axis. This mea…
https://en.wikipedia.org/wiki/Triple_helix

+ 3 more evidence sources

help

Claim 4: “Zihan Zhou et al, HELIX: a scalable model for predicting context-dependent regulation of RNA splicing and isoform usage, Nature Computational Science (2026). DOI: 10.1038/s43588-026-00988-w”

INSUFFICIENT EVIDENCE

No evidence was found in the provided search results to verify the specific author, title, date (2026), or DOI of the paper.

info

Claim 5: “The framework—Hierarchical Explainable LSTM for Isoform eXpression (HELIX)—overcomes the limitations of conventional approaches via a two-layer deep-learning architecture.”

SINGLE SOURCE

The name 'HELIX' and its purpose are mentioned in one web search result, but the specific 'two-layer deep-learning architecture' is not corroborated by a second independent source.

menu_book

wikipedia NEUTRAL — In machine learning, deep learning (DL) focuses on utilizing multilayered neural networks to perform tasks such as classification, regression, and representation learning. The field takes inspiration …
https://en.wikipedia.org/wiki/Deep_learning

menu_book

wikipedia NEUTRAL — Generative artificial intelligence (GenAI) is a subfield of artificial intelligence (AI) that uses generative models to generate text, images, videos, audio, software code (vibe coding) or other forms…
https://en.wikipedia.org/wiki/Generative_AI

menu_book

wikipedia NEUTRAL — A large language model (LLM) is a neural network trained on a vast amount of text for natural language processing tasks, especially language generation. LLMs can generate, summarize, translate and par…
https://en.wikipedia.org/wiki/Large_language_model

+ 3 more evidence sources

info

Claim 6: “using large colorectal cancer cohorts, the researchers identified widespread splicing dysregulation and abnormal isoform usage in tumor cells.”

SINGLE SOURCE

One web search result confirms that researchers used large colorectal cancer cohorts to identify splicing dysregulation and abnormal isoform usage using the framework.

travel_explore

web search NEUTRAL — Associations between the gut microbiome and fatigue in cancer patients.
https://www.tandfonline.com/doi/abs/10.1080/15567036.2020.18…

travel_explore

web search NEUTRAL — For example, using large colorectal cancer cohorts, the researchers identified widespread splicing dysregulation and abnormal isoform usage in tumor cells.
https://www.brightsurf.com/news/8OMP43N1/ai-driven-framework…

travel_explore

web search NEUTRAL — Cancer cells exhibited increased transcript complexity, with widespread 3'-UTR shortening and reduced intron retention.
https://journal.hep.com.cn/pac/EN/10.1093/procel/pwaf049

info

Claim 7: “The model was trained and optimized on large-scale short- and long-read RNA-seq datasets covering 30 distinct human tissues”

SINGLE SOURCE

A web search result explicitly states the model was trained on short- and long-read RNA-seq datasets covering 30 distinct human tissues, but this is the only source providing this specific detail.

menu_book

wikipedia NEUTRAL — RNA-Seq is a technique that allows transcriptome studies (see also Transcriptomics technologies) based on next-generation sequencing technologies. This technique is largely dependent on bioinformatics…
https://en.wikipedia.org/wiki/List_of_RNA-Seq_bioinformatics…

menu_book

wikipedia NEUTRAL — The signal recognition particle RNA, (also known as 7SL, 6S, ffs, or 4.5S RNA) is part of the signal recognition particle (SRP) ribonucleoprotein complex. SRP recognizes the signal peptide and binds t…
https://en.wikipedia.org/wiki/Signal_recognition_particle_RN…

+ 3 more evidence sources

info

Claim 8: “the team also developed scHELIX, a single-cell extension of HELIX specifically tailored for single-cell RNA sequencing data.”

SINGLE SOURCE

The provided evidence for claim 7 discusses other RNA-Seq tools (sm-PORE-cupine, IRISeq) but contains no mention of 'scHELIX'.

travel_explore

web search NEUTRAL — Linking RNA Structure to Cell Behaviour. Using sm-PORE-cupine, the researchers observed that RNA molecules can adopt different structures, and that these differences are linked to how efficiently prot…
https://www.rna-seqblog.com/astar-scientists-develop-new-met…

travel_explore

web search NEUTRAL — The discovery of the structure of the DNA double helix was one of the most important of the 20th century. In this educational video, explore Watson and Crick...
https://www.youtube.com/watch?v=1vm3od_UmFg

travel_explore

web search NEUTRAL — Researchers develop IRISeq and EnrichSci to map cellular neighborhoods and analyze rare aging cells in the brain simultaneously.
https://neurosciencenews.com/single-cell-genomics-aging-brai…

verified

Claim 9: “The study, which was published in Nature Computational Science on May 19”

VERIFIED BY REFERENCE

The provided evidence for claim 1 consists of generic study tools and Wikipedia entries on alpha helices; there is no mention of a publication in Nature Computational Science on May 19.

menu_book

wikipedia NEUTRAL — An alpha helix (or α-helix) is a sequence of amino acids in a protein that are twisted into a coil (a helix). The alpha helix is the most common structural arrangement in the secondary structure of pr…
https://en.wikipedia.org/wiki/Alpha_helix

menu_book

wikipedia NEUTRAL — Natural computing, also called natural computation, is a terminology introduced to encompass three classes of methods: 1) those that take inspiration from nature for the development of novel problem-s…
https://en.wikipedia.org/wiki/Natural_computing

menu_book

wikipedia NEUTRAL — In molecular biology, the double helix is the structure formed by double-stranded molecules of nucleic acids such as DNA. The double-helical structure of a nucleic acid complex arises as a consequence…
https://en.wikipedia.org/wiki/Nucleic_acid_double_helix

+ 3 more evidence sources

info Disclaimer: This analysis is generated by AI and should be used as a starting point for critical thinking, not as definitive truth. Claims are verified against publicly available sources. Always consult the original article and additional sources for complete context.