Base by Base

Goldberg DC et al., Cell Genomics - This episode examines the methylation screening array (MSA), a compact Infinium BeadChip optimized for trait-associated and cell-type CpGs and compatible with matched 5hmC profiling to enable scalable EWAS, cell deconvolution, and aging analyses. Key terms: DNA methylation, 5hmC, Infinium array, EWAS, epigenetic clock.

Study Highlights:
The authors developed the MSA to enable high-throughput profiling of ternary-code cytosine modifications (5mC, 5hmC, and unmodified C) across many human samples. They generated matched 5mC/5hmC atlases across multiple tissues and immune cell types and validated array accuracy against sequencing standards. Analyses show 5hmC is enriched at enhancers and gene bodies, complements 5mC in defining tissue identity, correlates positively with gene expression, is diluted in proliferative tissues, and contributes to epigenetic clock signals.

Conclusion:
The MSA is an accurate, reproducible, and scalable platform that maps ternary-code methylation across human tissues and highlights complementary roles of 5hmC and 5mC in tissue identity, aging, and trait-associated signals, enabling population-scale methylation screening.

Music:
Enjoy the music based on this article at the end of the episode.

Article title:
Scalable screening of ternary-code DNA methylation dynamics associated with human traits

First author:
Goldberg DC

Journal:
Cell Genomics

DOI:
10.1016/j.xgen.2025.100929

Reference:
Goldberg DC, Cloud C, Lee S.M., et al. Scalable screening of ternary-code DNA methylation dynamics associated with human traits. Cell Genomics. 2025;5:100929. https://doi.org/10.1016/j.xgen.2025.100929

License:
This episode is based on an open-access article published under the Creative Commons Attribution 4.0 International License (CC BY 4.0) – https://creativecommons.org/licenses/by/4.0/

Support:
Base by Base is independent and ad-free — no sponsors, no paywall. If an episode was worth your time, chip in and keep the papers audited and the original songs coming:
❤️ Support monthly: https://buy.stripe.com/cNifZhclVebvagk2JDgEg01
☕ One-time donation: https://donate.stripe.com/7sY4gz71B2sN3RWac5gEg00
More at basebybase.com

On PaperCast Base by Base you'll discover the latest in genomics, functional genomics, structural genomics, and proteomics.

Episode link: https://basebybase.com/episodes/papercast-base-by-base-70-encoding-corrupted

QC:
This episode was checked against the original article PDF and publication metadata for the episode release published on 2025-07-09.

QC Scope:
- article metadata and core scientific claims from the narration
- excludes analogies, intro/outro, and music
- transcript coverage: Audited the transcript’s coverage of MSA design and targets, 5mC/5hmC/3-state methylation, the bACE sequencing approach, tissue specificity and aging clocks, and blood-based cell deconvolution, plus study limitations.
- transcript topics: Ternary-code DNA methylation (5mC, 5hmC, unmodified C); MSA design and probe targeting; bACE protocol and 5hmC discrimination; Tissue-specific 5hmC/5mC distributions; 5hmC aging dynamics and epigenetic clocks; Blood-based immune cell deconvolution and age/sex effects

QC Summary:
- factual score: 10/10
- metadata score: 10/10
- supported core claims: 7
- claims flagged for review: 0
- metadata checks passed: 4
- metadata issues found: 0

Metadata Audited:
- article_doi
- article_title
- article_journal
- license

Factual Items Audited:
- Existence of a tri-state DNA methylation encoding (5mC, 5hmC, unmodified cytosine)
- MSA design comprises 284,317 unique probes across 269,094 genomic loci, with 145,426 loci overlapping EPICv2
- bACE protocol enables 5hmC profiling by combining bisulfite conversion with APOBEC3A deamination to distinguish 5mC, 5hmC, and C via dual readouts
- 5hmC is tissue-specific and enriched in CNS, with the lowest levels in colon and lymph nodes; 5hmC levels correlate positively with tissue-specific gene expression and are diluted
- MSA enables high-resolution cell-type deconvolution; 174 unique cell types are represented (including 82 brain cell types); 64 whole-blood methylomes show age- and sex-related immu
- 5hmC dynamics contribute to aging signals and epigenetic clocks; 5hmC alone can predict chronological age with accuracy comparable to 5mC-based clocks

QC result: Pass.

Cunningham C et al., Cell Genomics - This study used orthotopic breast PDX models, pooled and arrayed CRISPR/Cas9 screens, and Direct‑Capture Perturb‑seq to search for synthetic‑dosage‑lethal (SDL) partners of PLK1 across heterogeneous tumors. IGF2BP2 emerged as a top SDL hit using independent functional genomics approaches. Pharmacologic and genetic inhibition of IGF2BP2 impaired expansion of PLK1‑overexpressing tumors and altered stability of target mRNAs, supporting a mechanistic link. The work highlights a potential therapeutic strategy that targets a common vulnerability in PLK1‑high cancer cells despite intratumor heterogeneity. Key terms: PLK1, IGF2BP2, synthetic dosage lethality, CRISPR screen, breast PDX.

Study Highlights:
The authors combined pooled genome‑wide and arrayed CRISPR/Cas9 screens in breast PDX models with single‑cell Direct‑Capture Perturb‑seq to identify SDL interactions with PLK1. IGF2BP2 was repeatedly prioritized as a top SDL hit across independent screens. Functional follow‑up showed that inhibition of IGF2BP2 decreased expansion of PLK1‑overexpressing tumors and destabilized mRNAs linked to the PLK1 state. The results point to IGF2BP2 as a mechanistic and potentially druggable dependency in PLK1‑high tumors.

Conclusion:
Across multiple orthogonal in vivo and single‑cell genomic screens, IGF2BP2 is identified as a reproducible synthetic‑dosage‑lethal partner of PLK1; targeting IGF2BP2 reduces growth of PLK1‑overexpressing breast PDX tumors by perturbing mRNA stability pathways and may offer a strategy to exploit a shared vulnerability in heterogeneous cancers.

Music:
Enjoy the music based on this article at the end of the episode.

Article title:
Three independent approaches identify IGF2BP2 as a top SDL target of PLK1

First author:
Cunningham C

Journal:
Cell Genomics

DOI:
10.1016/j.xgen.2025.100876

Reference:
Cunningham C.E., Vizeacoumar F.S., Zhang Y., Kyrylenko L., Both S., Maranda V., et al.. Identification of targetable vulnerabilities of PLK1-overexpressing cancers by synthetic dosage lethality. Cell Genomics, 5, 100876. (2025). https://doi.org/10.1016/j.xgen.2025.100876

License:
This episode is based on an open-access article published under the Creative Commons Attribution 4.0 International License (CC BY 4.0) – https://creativecommons.org/licenses/by/4.0/

Support:
Base by Base – Stripe donations: https://donate.stripe.com/7sY4gz71B2sN3RWac5gEg00

Official website https://basebybase.com

On PaperCast Base by Base you'll discover the latest in genomics, functional genomics, structural genomics, and proteomics.

Episode link: https://basebybase.com/episodes/plk1-igf2bp2-sdl-69

QC:
This episode was checked against the original article PDF and publication metadata for the episode release published on 2025-07-08.

QC Scope:
- article metadata and core scientific claims from the narration
- excludes analogies, intro/outro, and music
- transcript coverage: Audited sections describing SDL concept, IGF2BP2 mechanism (PLK1 mRNA stability and OxPhos transcripts), Direct-Capture Perturb-seq validation, pharmacological inhibition with C4, in vivo breast cancer PDX findings, safety window vs normal cells, and study limitations.
- transcript topics: SDL-based identification of IGF2BP2 as PLK1 dependency; IGF2BP2's role in PLK1 mRNA stability; IGF2BP2's stabilization of OxPhos transcripts; Direct-Capture Perturb-seq validation across heterogeneous tumors; Pharmacological IGF2BP2 inhibition with compound C4; Selective toxicity and therapeutic window in normal vs cancer cells

QC Summary:
- factual score: 10/10
- metadata score: 10/10
- supported core claims: 8
- claims flagged for review: 0
- metadata checks passed: 4
- metadata issues found: 0

Metadata Audited:
- article_doi
- article_title
- article_journal
- license

Factual Items Audited:
- PLK1 is overexpressed in rapidly dividing cancer cells across tumor types
- IGF2BP2 identified as a top SDL hit with PLK1 by three independent approaches
- IGF2BP2 binds to PLK1 mRNA and stabilizes it; IGF2BP2 inhibition leads to downregulation of PLK1 transcripts
- IGF2BP2 also stabilizes transcripts driving oxidative phosphorylation (OXPHOS)
- Inhibition of IGF2BP2 reduces expansion of PLK1-overexpressing tumors and destabilizes target transcripts
- Exogenous ATP partially rescues energy-deprived cancer cells, linking lethality to metabolic collapse

QC result: Pass.

Tenthorey JL et al., Cell Genomics - This episode examines a study showing that insertion/deletion mutations (indels) in the v1 loop of the antiviral protein TRIM5a can create new viral specificities in a single step, whereas missense mutations often cannot. The authors used saturation missense mutagenesis, combinatorial libraries, and a novel deep indel scanning approach to compare evolutionary potential. Key terms: TRIM5a, indels, deep mutational scanning, SIVsab, antiviral evolution.

Study Highlights:
The authors performed saturation missense mutagenesis and a novel deep indel scanning of the TRIM5a v1 loop to compare mutational routes to new antiviral specificity. No single missense mutation enabled human TRIM5a to restrict SIVsab, and combinatorial sampling showed six specific changes were required to recapitulate rhesus-like activity. In contrast, a single in-frame duplication of phenylalanine (F339dup) conferred potent SIVsab restriction and broadened activity to other lentiviruses. Naturally occurring primate TRIM5a indels similarly confer novel specificities, demonstrating indels as a distinct source of evolutionary innovation.

Conclusion:
In-frame indel mutations in disordered virus-binding loops can enable rapid, large-effect gains in antiviral function that are inaccessible to single amino-acid substitutions, and such indels have contributed to TRIM5a evolution across primates.

Music:
Enjoy the music based on this article at the end of the episode.

Article title:
Indels allow antiviral proteins to evolve functional novelty inaccessible by missense mutations

First author:
Tenthorey JL

Journal:
Cell Genomics

DOI:
10.1016/j.xgen.2025.100818

Reference:
Tenthorey JL, del Banco S, Ramzan I, Klingenberg H, Liu C, Emerman M, Malik HS. Indels allow antiviral proteins to evolve functional novelty inaccessible by missense mutations. Cell Genomics. 2025;5:100818. doi:10.1016/j.xgen.2025.100818

License:
This episode is based on an open-access article published under the Creative Commons Attribution 4.0 International License (CC BY 4.0) – https://creativecommons.org/licenses/by/4.0/

Support:
Base by Base – Stripe donations: https://donate.stripe.com/7sY4gz71B2sN3RWac5gEg00

Official website https://basebybase.com

On PaperCast Base by Base you'll discover the latest in genomics, functional genomics, structural genomics, and proteomics.

Episode link: https://basebybase.com/episodes/indels-antiviral-evolution-trim5a-68

QC:
This episode was checked against the original article PDF and publication metadata for the episode release published on 2025-07-07.

QC Scope:
- article metadata and core scientific claims from the narration
- excludes analogies, intro/outro, and music
- transcript coverage: Substantive auditing of TRIM5a v1 loop mutagenesis results, missense vs. indel contrasts, the F339dup indel finding, and the evolutionary history of indels in primates, including biophysical interpretation and limitations.
- transcript topics: TRIM5a v1 loop and lentiviral restriction; Missense mutational scanning (DMS) vs indel mutational potential; Deep indel scanning (DIS) and the F339dup finding; Natural primate TRIM5a indels and evolutionary history; Biophysical interpretation: entropic penalty and loop pre-folding; Evolutionary implications for antiviral specificity

QC Summary:
- factual score: 10/10
- metadata score: 10/10
- supported core claims: 3
- claims flagged for review: 0
- metadata checks passed: 4
- metadata issues found: 0

Metadata Audited:
- article_doi
- article_title
- article_journal
- license

Factual Items Audited:
- Missense mutations in the TRIM5a v1 loop do not confer SIVsab restriction in human TRIM5a
- A single in-frame indel (F339dup) enables potent SIVsab restriction in human TRIM5a in one evolutionary step
- F339dup also broadens restriction to HIV-1 and SIVcpz without apparent cost
- A 5xRh plus a sixth rhesus-mimicking mutation is required for SIVsab restriction in human TRIM5a via missense mutations; six independent mutations are needed
- Naturally occurring primate TRIM5a indels (2-aa insertion; 20-aa duplication; 9-aa deletion) map to evolving antiviral specificities

QC result: Pass.

Zhou Y et al., The American Journal of Human Genetics - This episode explores M-REGLE, a multimodal deep‑learning pipeline that jointly learns representations from ECG and PPG waveforms to boost GWAS discovery and polygenic risk prediction for cardiovascular traits, including atrial fibrillation, and validates results across multiple biobanks. Key terms: M-REGLE, multimodal learning, ECG, PPG, GWAS.

Study Highlights:
The authors developed M-REGLE, an early‑fusion convolutional variational autoencoder that jointly encodes complementary electrophysiological waveforms (12‑lead ECG, ECG lead I, and PPG) into low‑dimensional embeddings, then ran GWAS on orthogonalized PCs and combined chi‑squared statistics. Compared to unimodal representation learning (U‑REGLE), M‑REGLE reduced reconstruction error, discovered more genome‑wide significant hits and loci (e.g., ~19.3% more loci on 12‑lead ECG and ~13.0% more on ECG lead I + PPG), and yielded higher expected chi‑squared statistics. Polygenic risk scores built from M‑REGLE hits significantly improved prediction for several cardiovascular phenotypes, notably atrial fibrillation, and findings were validated in Indiana Biobank, EPIC‑Norfolk, and the British Women’s Heart and Health Study. The method generalized to adding a third modality (spirograms) and outperformed PCA and CAE baselines in power and enrichment for cardiovascular terms.

Conclusion:
Joint multimodal representation learning with M‑REGLE leverages complementary ECG and PPG signals to increase GWAS power and improve polygenic risk prediction for cardiac traits, demonstrating a practical route to use wearable waveform data for genetic discovery.

Music:
Enjoy the music based on this article at the end of the episode.

Article title:
Applying multimodal AI to physiological waveforms improves genetic prediction of cardiovascular traits

First author:
Zhou Y

Journal:
The American Journal of Human Genetics

DOI:
10.1016/j.ajhg.2025.05.015

Reference:
Zhou Y., Khasentino J., Yun T., et al. Applying multimodal AI to physiological waveforms improves genetic prediction of cardiovascular traits. The American Journal of Human Genetics. 2025;112:1562–1579. https://doi.org/10.1016/j.ajhg.2025.05.015

License:
This episode is based on an open-access article published under the Creative Commons Attribution 4.0 International License (CC BY 4.0) – https://creativecommons.org/licenses/by/4.0/

Support:
Base by Base – Stripe donations: https://donate.stripe.com/7sY4gz71B2sN3RWac5gEg00

Official website https://basebybase.com

On PaperCast Base by Base you'll discover the latest in genomics, functional genomics, structural genomics, and proteomics.

Episode link: https://basebybase.com/episodes/m-regle-multimodal-ai-ecg-ppg-gwas

QC:
This episode was checked against the original article PDF and publication metadata for the episode release published on 2025-07-06.

QC Scope:
- article metadata and core scientific claims from the narration
- excludes analogies, intro/outro, and music
- transcript coverage: Audited sections cover the M-REGLE architecture (early fusion VAE), comparison with unimodal REGLE, reconstruction improvements, GWAS power gains, PRS performance for atrial fibrillation, cross-biobank validation, and extension to spirograms.
- transcript topics: M-REGLE concept and early fusion; VAE latent embeddings and joint multimodal representation; Canonical correlation between ECG lead I and PPG; GWAS power gains and loci discovery (12-lead ECG and lead I + PPG); Polygenic risk scores and atrial fibrillation prediction; Cross-biobank validation (UK Biobank, Indiana Biobank, EPIC-Norfolk, BWHHS)

QC Summary:
- factual score: 10/10
- metadata score: 10/10
- supported core claims: 4
- claims flagged for review: 0
- metadata checks passed: 4
- metadata issues found: 0

Metadata Audited:
- article_doi
- article_title
- article_journal
- license

Factual Items Audited:
- Early fusion M-REGLE combines ECG and PPG data into a single joint representation for GWAS and downstream analyses.
- M-REGLE reduces waveform reconstruction error compared to unimodal methods (72.5% reduction for 12-lead ECG; 20.2% reduction for ECG lead I + PPG).
- M-REGLE identifies more genetic loci than unimodal approaches: 262 hits and 142 loci on 12-lead ECG, with 20 novel loci; ECG lead I + PPG yields 103 hits and 61 loci with 14 and 7
- There is a 22.0% increase in the expected chi-squared statistic for 12-lead ECG when using M-REGLE versus unimodal methods (and 16.4% for ECG lead I + PPG).
- Polygenic risk scores built from M-REGLE hits outperform unimodal PRS in predicting atrial fibrillation in UK Biobank (AUROC 0.59 vs 0.57; AUPRC 0.10 vs 0.09).
- PRS performance for Afib and related phenotypes validated in independent biobanks (Indiana Biobank, EPIC-Norfolk, British Women’s Heart and Health Study).

QC result: Pass.

🎙️ Episode 67: Mainstreaming Clinical Genetic Testing — A Conceptual Framework

🧬 In this episode of Base by Base, we explore a unified approach to integrating genetic testing into routine patient care, as proposed by Mackley et al. (2025) in Genetics in Medicine. Drawing on a national Canadian focus group of clinical genetics experts, the authors establish consensus definitions for key terms such as “mainstreaming,” “diagnostic genetic test,” and “genetics service,” then map the diagnostic care pathway into four stages—assessment, pre-testing, laboratory, and post-testing—and identify the variables that determine how responsibility shifts between non-geneticist clinicians and genetics teams .

🔍 Study highlights: The authors clarify foundational terminology and outline the full set of activities in the genetic testing process; they identify six categories of variables—patient, disease, test, clinician, report, and health-system characteristics—that influence each scenario’s readiness for mainstreaming; they propose four distinct models of care (from “to-test,” where genetics services arrange testing based on referral details, through “to-navigation,” where non-geneticist clinicians manage nearly all steps, to a fully autonomous model); they provide a decision tree guiding when genetics services should re-engage under the “to-navigation” model; and they emphasize the need for standardized implementation and evaluation to enable cross-program comparisons.

🧠 Conclusion: By offering a granular yet generalizable roadmap, this framework empowers healthcare teams to design, assess, and refine mainstreaming strategies, fostering effective partnerships between genetics specialists and front-line providers to optimize resources and improve patient access to genomic medicine.

📖 Reference: Mackley MP, Richer J, Guerin A, et al. Mainstreaming of clinical genetic testing: a conceptual framework. Genetics in Medicine. 2025. https://doi.org/10.1016/j.gim.2025.101465

📜 License: This episode is based on an open access article published under the Creative Commons Attribution 4.0 International License (CC BY 4.0) – https://creativecommons.org/licenses/by/4.0/

Mackley MP et al., Genetics in Medicine - This episode summarizes a consensus-derived framework for mainstreaming clinical genetic testing developed from a Canadian expert focus group. The framework defines terminology, maps diagnostic pathway activities, and describes four models that vary by when genetics services become involved. Key terms: Clinical genetics, Genetic testing, Mainstreaming, Service delivery, Genomic medicine.

Study Highlights:
The authors convened a consensus focus group of 35 genetics experts from 20 clinical genetics services across Canada and combined this with literature review and iterative manuscript revision. They produced consensus definitions for mainstreaming-related terms and grouped diagnostic activities into four stages. The framework defines four mainstreaming models (to-test, to-result, to-navigation, autonomous) distinguished by the point at which responsibility shifts from nongeneticist clinicians to genetics services and identifies six categories of variables that influence model suitability. The authors propose the framework to guide standardized design, reporting, and evaluation of mainstreaming efforts to optimize genetics resources and patient care.

Conclusion:
A generalizable, high-level framework describing four mainstreaming models and influencing variables can guide implementation and standardized evaluation of clinical genetic testing outside traditional genetics services to improve access and resource use.

Music:
Enjoy the music based on this article at the end of the episode.

Article title:
Mainstreaming of clinical genetic testing: A conceptual framework

First author:
Mackley MP

Journal:
Genetics in Medicine

DOI:
10.1016/j.gim.2025.101465

Reference:
Mackley MP, Richer J, Guerin A, et al. Mainstreaming of clinical genetic testing: A conceptual framework. Genetics in Medicine. 2025;27:101465. doi:https://doi.org/10.1016/j.gim.2025.101465

License:
This episode is based on an open-access article published under the Creative Commons Attribution 4.0 International License (CC BY 4.0) – https://creativecommons.org/licenses/by/4.0/

Support:
Base by Base – Stripe donations: https://donate.stripe.com/7sY4gz71B2sN3RWac5gEg00

Official website https://basebybase.com

On PaperCast Base by Base you'll discover the latest in genomics, functional genomics, structural genomics, and proteomics.

Episode link: https://basebybase.com/episodes/mainstreaming-clinical-genetic-testing-framework

QC:
This episode was checked against the original article PDF and publication metadata for the episode release published on 2025-07-05.

QC Scope:
- article metadata and core scientific claims from the narration
- excludes analogies, intro/outro, and music
- transcript coverage: Audited the transcript sections that define the mainstreaming concept, describe the four models, enumerate the six influence variables, map models to the diagnostic care pathway, provide real-world examples, discuss reporting readability and workforce education, and cover limitations and cascade-screening implications.
- transcript topics: Mainstreaming concept and framework; Four mainstreaming models: to-test, to-result, to-navigation, autonomous; Six variables influencing model suitability; Mapping models to diagnostic care pathway (assessment, pretesting, laboratory, post-testing); Real-world examples (CMA in autism, oncology panels, lab reporting/readability, cascade screening); Lab reporting readability and clinician education

QC Summary:
- factual score: 10/10
- metadata score: 10/10
- supported core claims: 7
- claims flagged for review: 0
- metadata checks passed: 4
- metadata issues found: 0

Metadata Audited:
- article_doi
- article_title
- article_journal
- license

Factual Items Audited:
- Four mainstreaming models described: to-test, to-result, to-navigation, autonomous
- Six influencing variables for model suitability: patient, disease, test, report, clinician, system
- Models mapped to stages of the diagnostic care pathway: assessment, pretesting, laboratory, post-testing
- Real-world examples linked to models (CMA in autism, oncology panels, lab reporting/readability, cascade screening)
- Emphasis on standardization, evaluation, and workforce education in genomics
- Limitations acknowledged: Canadian context and representation gaps

QC result: Pass.

Walker S et al., Genetics in Medicine - This episode reviews a systematic analysis of ultra-rare FBN1 variants in the 100,000 Genomes Project using SpliceAI, RNA assays and minigene tests. The study identified 20 non-canonical splice variants across 23 families, confirmed splicing defects for 16 variants, and estimates these variants account for ~3% of undiagnosed FTAAD/Marfan families. The work highlights the value of intronic analysis and confirmatory RNA testing in clinical genomics. Key terms: Marfan syndrome, FBN1, splicing, pseudoexon, genome sequencing.

Study Highlights:
The authors screened 78,195 genomes from the 100,000 Genomes Project and identified 20 unique non-canonical FBN1 splice variants in 23 families, with significant enrichment among participants recruited for Familial Thoracic Aortic Aneurysm Disease (OR=84, p=9.7x10-14). Experimental validation (RT-PCR, RNAseq, minigene assays) confirmed aberrant splicing in 16 of 20 variants and revealed pseudoexonization as a common mechanism. Most actionable variants lay beyond the ±8 bp windows used in routine testing, and the study estimates a diagnostic yield of ~3% from these non-canonical splice events in undiagnosed FTAAD families.

Conclusion:
Genome sequencing combined with expanded splice prediction and confirmatory RNA testing uncovers non-canonical FBN1 splice variants that meaningfully increase molecular diagnoses for Marfan/FTAAD and should be integrated into testing pipelines with appropriate RNA-capacity.

Music:
Enjoy the music based on this article at the end of the episode.

Article title:
Utility of genome sequencing and group-enrichment to support splice variant interpretation in Marfan syndrome

First author:
Walker S

Journal:
Genetics in Medicine

DOI:
10.1016/j.gim.2025.101477

Reference:
Walker S, Bunyan DJ, Thomas HB, et al. Utility of genome sequencing and group-enrichment to support splice variant interpretation in Marfan syndrome. Genetics in Medicine (2025). doi: https://doi.org/10.1016/j.gim.2025.101477

License:
© 2025 Published by Elsevier Inc. on behalf of American College of Medical Genetics and Genomics.

Support:
Base by Base – Stripe donations: https://donate.stripe.com/7sY4gz71B2sN3RWac5gEg00

Official website https://basebybase.com

On PaperCast Base by Base you'll discover the latest in genomics, functional genomics, structural genomics, and proteomics.

Episode link: https://basebybase.com/episodes/fibrillin1-noncanonical-splice-marfan-100kgp

QC:
This episode was checked against the original article PDF and publication metadata for the episode release published on 2025-07-04.

QC Scope:
- article metadata and core scientific claims from the narration
- excludes analogies, intro/outro, and music
- transcript coverage: Substantive audit focused on the deep intronic FBN1 splice-variant discovery, computational prediction with expanded windows, laboratory validation (RNA, minigene), enrichment statistics, diagnostic yield, and ACMG interpretation; also notes limitations and replication in UK Biobank.
- transcript topics: Non-canonical FBN1 splice variants in 100kGP; SpliceAI prediction and 500bp window expansion; RNA validation: RT-PCR and RNAseq; Minigene assays and functional testing; Enrichment in FTAAD and odds ratio (OR); Diagnostic yield and cascade testing implications

QC Summary:
- factual score: 10/10
- metadata score: 10/10
- supported core claims: 6
- claims flagged for review: 0
- metadata checks passed: 4
- metadata issues found: 0

Metadata Audited:
- article_doi
- article_title
- article_journal
- license

Factual Items Audited:
- 20 unique cryptic FBN1 splice variants identified across 23 families (32 affected individuals)
- 7:14 split? (clarify not needed) 70% of variants lie beyond the ±8 splice regions
- RNA or functional confirmation performed for 16/20 variants
- 9/20 variants predicted to cause pseudoexon inclusion
- Odds ratio for enrichment of candidate splice variants in FTAAD cohort: OR = 84 (p = 9.7e-14)
- Diagnostic yield estimated at ~3% among unsolved FTAAD families

QC result: Pass.

Xu H et al., Cell Genomics - This episode covers Pisces, a machine-learning framework that augments sparse drug-pair datasets by creating multiple modality-based views per drug to improve prediction of drug synergy, xenograft responses, and drug-drug interactions. Key terms: drug combination, data augmentation, multimodal, machine learning, drug-drug interaction.

Study Highlights:
The authors introduce Pisces, which embeds eight drug modalities and combines them pairwise to expand each drug-pair into up to 64 training views, allowing the model to handle missing modalities. Pisces applies contrastive alignment, noisy-label top-k aggregation, and ResNet integration to produce robust predictions. The method attains state-of-the-art performance on cell-line (GDSC-combo), xenograft tumor response, and drug–drug interaction benchmarks. Interpreting Pisces predictions with a genetic interaction network revealed a BRCA drug-sensitive pathway validated in TCGA with survival differences.

Conclusion:
Multi-modal data augmentation by treating modality pairs as independent views enables Pisces to substantially improve prediction of drug synergy, xenograft tumor response, and DDIs and to generate biologically and clinically relevant hypotheses such as a BRCA drug-sensitive pathway.

Music:
Enjoy the music based on this article at the end of the episode.

Article title:
Pisces: A multi-modal data augmentation approach for drug combination synergy prediction

First author:
Xu H

Journal:
Cell Genomics

DOI:
10.1016/j.xgen.2025.100892

Reference:
Xu H., Lin J., Woicik A., Liu Z., Ma J., Zhang S., Poon H., Wang L., Wang S. Pisces: A multi-modal data augmentation approach for drug combination synergy prediction. Cell Genomics. 2025;5:100892. https://doi.org/10.1016/j.xgen.2025.100892

License:
This episode is based on an open-access article published under the Creative Commons Attribution 4.0 International License (CC BY 4.0) – https://creativecommons.org/licenses/by/4.0/

Support:
Base by Base is independent and ad-free — no sponsors, no paywall. If an episode was worth your time, chip in and keep the papers audited and the original songs coming:
❤️ Support monthly: https://buy.stripe.com/cNifZhclVebvagk2JDgEg01
☕ One-time donation: https://donate.stripe.com/7sY4gz71B2sN3RWac5gEg00
More at basebybase.com

On PaperCast Base by Base you'll discover the latest in genomics, functional genomics, structural genomics, and proteomics.

Episode link: https://basebybase.com/episodes/base-by-base-64-garbled-pdf

QC:
This episode was checked against the original article PDF and publication metadata for the episode release published on 2025-07-03.

QC Scope:
- article metadata and core scientific claims from the narration
- excludes analogies, intro/outro, and music
- transcript coverage: Audited transcript sections describing Pisces architecture, data augmentation (64 views), the noisy-aggregation approach, key experimental results (cell lines, unseen drug pairs, xenografts, DDI), BRCA pathway/TCGA validation, and stated limitations.
- transcript topics: Problem of drug synergy prediction and data scarcity; Pisces multimodal data augmentation concept; Eight modalities per drug and 64 augmented views; Projector, augmentor, and aggregator architecture; Noisy label learning and top-k aggregation; Evaluation on GDSC cell lines (unseen drug combos, unseen cell lines)

QC Summary:
- factual score: 10/10
- metadata score: 10/10
- supported core claims: 7
- claims flagged for review: 0
- metadata checks passed: 4
- metadata issues found: 0

Metadata Audited:
- article_doi
- article_title
- article_journal
- license

Factual Items Audited:
- Pisces expands data by 64× by using eight modalities per drug and pairwise combination, producing 64 augmented views for a drug pair.
- Pisces architecture consists of projector, augmentor, and aggregator, with top-k aggregation to form final predictions.
- Unseen two-drug combinations show F1 improvement of about 24% over the next-best approach; unseen cell lines show F1 improvement >10%.
- Three-drug synergy prediction achieved AUROC of 0.8525 when trained only on two-drug data.
- BRCA drug-sensitive pathway consisting of 19 genes was identified by Pisces and validated with TCGA patient survival data.
- Pisces can bypass missing modalities by skipping pairs that require missing views, avoiding imputation.

QC result: Pass.

Steiner MC et al., PNAS - This episode examines a theoretical and empirical study showing how the geographic breadth of sampling affects discovery and observed frequencies of deleterious rare variants. The authors develop a spatial stochastic model, validate it with simulations, and test predictions using UK Biobank exome resampling. Key terms: sampling breadth, rare variants, negative selection, site frequency spectrum, UK Biobank.

Study Highlights:
The authors build a stochastic spatial model that combines dispersal, genetic drift, selection, mutation, and geographically concentrated sampling to derive expected sample site frequency spectra for deleterious variants. They find that broader sampling increases the number of distinct variants discovered ("discovery") while reducing the average observed frequency per variant ("dilution"). The magnitude of these effects scales with the sampling breadth relative to the allele spread length (`c), sample size, and selection strength. Theoretical predictions are validated with branching-process and SLiM simulations and by in silico resampling of UK Biobank exomes.

Conclusion:
Geographic sampling breadth produces a trade-off: broader samples discover more distinct deleterious variants but each at lower frequency, a pattern that affects association study power and SFS-based inference of negative selection. Study design for biobank-scale genetics should explicitly account for sampling breadth and its downstream effects.

Music:
Enjoy the music based on this article at the end of the episode.

Article title:
Study design and the sampling of deleterious rare variants in biobank-scale datasets

First author:
Steiner MC

Journal:
PNAS

DOI:
10.1073/pnas.2425196122

Reference:
Steiner MC, Rice DP, Biddanda A, Ianni-Ravna MK, Porras C, Novembre J. Study design and the sampling of deleterious rare variants in biobank-scale datasets. PNAS. 2025;122:e2425196122. https://doi.org/10.1073/pnas.2425196122

License:
This episode is based on an open-access article published under the Creative Commons Attribution 4.0 International License (CC BY 4.0) – https://creativecommons.org/licenses/by/4.0/

Support:
Base by Base – Stripe donations: https://donate.stripe.com/7sY4gz71B2sN3RWac5gEg00

Official website https://basebybase.com

On PaperCast Base by Base you'll discover the latest in genomics, functional genomics, structural genomics, and proteomics.

Episode link: https://basebybase.com/episodes/discovery-dilution-sampling-breadth-rare-variants

QC:
This episode was checked against the original article PDF and publication metadata for the episode release published on 2025-07-02.

QC Scope:
- article metadata and core scientific claims from the narration
- excludes analogies, intro/outro, and music
- transcript coverage: Audited the transcript portions describing (1) geographic breadth effects on variant discovery and frequency (discovery vs dilution), (2) the stochastic spatial model and sampling breadth parameter w, (3) UK Biobank exome resampling results, and (4) implications for GWAS power and inference of negative selection, inclu
- transcript topics: Geographic breadth and site frequency spectrum (SFS); Discovery and dilution trade-off; Stochastic spatial model and sampling breadth (w) and characteristic length (c); UK Biobank exome resampling results (LoF variants, heterozygosity, singletons); Implications for GWAS power and inference of negative selection; Model limitations and boundary conditions (torus assumption vs real geography)

QC Summary:
- factual score: 10/10
- metadata score: 10/10
- supported core claims: 7
- claims flagged for review: 0
- metadata checks passed: 4
- metadata issues found: 0

Metadata Audited:
- article_doi
- article_title
- article_journal
- license

Factual Items Audited:
- Broad geographic sampling increases the number of distinct deleterious variants discovered (discovery).
- Broad geographic sampling reduces the average frequency of each discovered variant (dilution).
- The average allele frequency across the entire study is unchanged by sampling breadth (discovery-dilution balance).
- UK Biobank resampling shows 72.3% more loss-of-function (LoF) variants discovered with broader sampling, and 36.75% lower heterozygosity at LoF variant sites; singletons increase s
- There is a discovery vs dilution trade-off affecting GWAS power and inference of negative selection.

QC result: Pass.

Pappas K et al., PNAS - A genome-wide CRISPR screen in Brca2‑deficient murine prostate organoids identifies loss-of-function in DNA prereplication complex genes (CDT1, CDC6, DBF4) as a reversion‑independent mechanism of resistance to PARP inhibitors; pharmacologic disruption of the Geminin–CDT1 interaction can restore sensitivity. Key terms: PARP inhibitors, BRCA2, prereplication complex, prostate cancer, organoids.

Study Highlights:
Using primary mouse prostate organoids with engineered Brca2 loss and genome-wide CRISPR screening, the authors found multiple independent sgRNAs targeting pre‑RC genes (Cdt1, Cdc6, Dbf4) that confer resistance to olaparib and the PARP1-selective inhibitor AZD5305. Pre‑RC loss promoted rapid resolution of PARPi‑induced DNA damage and rescued replication fork protection in Brca2‑deficient cells. Copy number loss of CDT1 is frequent in CRPC and overlaps with BRCA2 loss, suggesting potential biomarker relevance. Pharmacologic disruption of the Geminin–CDT1 interaction (AF615) restored PARPi sensitivity in selected pre‑RC deficient models.

Conclusion:
Impaired DNA prereplication complex function provides a reversion‑mutation‑independent route to PARP inhibitor resistance in BRCA2‑deficient prostate cancer models, and targeting the Geminin–CDT1 axis may reverse resistance in specific contexts.

Music:
Enjoy the music based on this article at the end of the episode.

Article title:
BRCA2 reversion mutation–independent resistance to PARP inhibition through impaired DNA prereplication complex function

First author:
Pappas K

Journal:
PNAS

DOI:
10.1073/pnas.2426743122

Reference:
Pappas K., Ferrari M., Smith P., et al. BRCA2 reversion mutation–independent resistance to PARP inhibition through impaired DNA prereplication complex function. PNAS. 2025;122(23):e2426743122. doi:10.1073/pnas.2426743122

License:
This episode is based on an open-access article published under the Creative Commons Attribution 4.0 International License (CC BY 4.0) – https://creativecommons.org/licenses/by/4.0/

Support:
Base by Base – Stripe donations: https://donate.stripe.com/7sY4gz71B2sN3RWac5gEg00

Official website https://basebybase.com

On PaperCast Base by Base you'll discover the latest in genomics, functional genomics, structural genomics, and proteomics.

Episode link: https://basebybase.com/episodes/pre-rc-loss-parpi-resistance

QC:
This episode was checked against the original article PDF and publication metadata for the episode release published on 2025-07-01.

QC Scope:
- article metadata and core scientific claims from the narration
- excludes analogies, intro/outro, and music
- transcript coverage: Audited sections describing BRCA2-deficient PARPi resistance, the DNA prereplication complex (CDT1, CDC6, DBF4), replication fork protection and γH2AX dynamics, Geminin regulation and AF615/Geminin–CDT1 axis as a therapeutic angle, and clinical biomarker implications (CDT1 copy-number loss in CRPC).
- transcript topics: BRCA2-deficient PARP inhibitor resistance in prostate organoids; DNA prereplication complex (CDT1, CDC6, DBF4) and PARPi resistance; Replication fork protection and γH2AX dynamics under PARPi; Geminin regulation of pre-RC and CDT1 sequestration; AF615 and AZD5305 combination therapy to reverse resistance; CDT1 copy-number loss as a CRPC biomarker

QC Summary:
- factual score: 10/10
- metadata score: 10/10
- supported core claims: 6
- claims flagged for review: 0
- metadata checks passed: 4
- metadata issues found: 0

Metadata Audited:
- article_doi
- article_title
- article_journal
- license

Factual Items Audited:
- Upfront PARP inhibitor resistance is common in BRCA-mutant CRPC (~50%).
- Genome-wide CRISPR screens repeatedly identified CDT1, CDC6, and DBF4 as resistance drivers (pre-RC genes).
- Impaired pre-RC function leads to rapid resolution of PARPi-induced DNA damage and restoration of fork protection.
- Geminin knockdown or AF615 (Geminin–CDT1 inhibitor) reverses pre-RC–related PARPi resistance; AF615 restores sensitivity when combined with AZD5305.
- CDT1 copy-number loss is common in CRPC and co-occurs with BRCA2 loss.
- In BRCA2-deficient organoids with pre-RC loss, in vivo xenografts retain CDT1 deletion and PARPi resistance, supporting in vivo relevance.

QC result: Pass.