Base by Base

Gustavo Barra
  • 5,0 (1)
  • Ciências da vida
  • Diário

Base by Base explores advances in genetics and genomics, with a focus on gene-disease associations, variant interpretation, protein structure, and insights from exome and genome sequencing. Each episode breaks down key studies and their clinical relevance—one base at a time. Powered by AI, Base by Base offers a new way to learn on the go. Special thanks to authors who publish under CC BY 4.0, making open-access science faster to share and easier to explore.

  1. há 6 h

    462: A vessel gene tied to AMD dims low-light vision in mice

    Cheng et al., Proceedings of the National Academy of Sciences - Genome-wide studies have found dozens of risk loci for age-related macular degeneration, but for many of them the gene doing the work is unknown. This study combines AMD genetics with blood and retinal gene-expression data to prioritize nine candidate genes, screens them in zebrafish, and follows the strongest, CNN2, into knockout mice. Mice lacking Cnn2 lose sensitivity in dim light and to fine contrast while acuity stays normal, and their photoreceptor layer thins, even though the gene is expressed mainly in the retina's blood-vessel cells rather than in photoreceptors. Key terms: age-related macular degeneration, CNN2, statistical genetics, photoreceptor degeneration, retinal neurovascular homeostasis. Study Highlights: Summary-data-based Mendelian randomization combining an AMD genome-wide study of 16,144 cases and 17,832 controls with blood expression data from 2,765 people found 16 genes, and the HEIDI test kept 9 whose signals were not explained by linkage, 4 of which replicated in retinal data from 406 donors. In zebrafish larvae, morpholino knockdown of cnn2 shrank eye area by 48.5% and axial length by 32.5%, cut the light-on visual motor response by 61.9%, and both defects were largely rescued by injecting cnn2 mRNA. Cnn2 knockout mice at 3 months had a 39.2% lower a-wave and a 27.3% lower b-wave in dim-light electroretinograms and needed 15.3% contrast to respond where wild-type mice needed 4.1%, while bright-light responses and visual acuity were unchanged. The photoreceptor layer thinned by 3 and 5 months, cone arrestin and rhodopsin fell, and the 436 genes altered in the retina were enriched 6.02-fold near known AMD risk loci. Cnn2 was expressed mainly in endothelial cells and pericytes, and in human retinas it was lower in AMD but not significantly so. Conclusion: Human genetic evidence and experiments in two species converge on CNN2, a calponin gene active mainly in retinal blood-vessel cells, as a candidate susceptibility gene for age-related macular degeneration. The authors propose that its loss disturbs neurovascular balance and harms photoreceptors secondarily, but the data do not yet show whether the effect is direct or indirect, and neither model has a macula. Music: Enjoy the music based on this article at the end of the episode. Article title: Genetic evidence and cross-species functional characterization implicate CNN2 in age-related macular degeneration susceptibility First author: Cheng Journal: Proceedings of the National Academy of Sciences DOI: 10.1073/pnas.2533682123 Reference: Cheng, F.-F., Mou, H., Liu, Z., Zhang, C.-J., Zhuang, Y.-Y., Wu, Z., Wen, X.-R., Xue, A., Zhang, X., Yang, J., and Jin, Z.-B. (2026). Genetic evidence and cross-species functional characterization implicate CNN2 in age-related macular degeneration susceptibility. Proceedings of the National Academy of Sciences 123(36), e2533682123. https://doi.org/10.1073/pnas.2533682123 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 ▶️ Watch with chapters on YouTube: https://www.youtube.com/@basebybase 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/cnn2-amd-vessel-gene-low-light-vision-mice QC: This... Chapters (00:00:10) - Intro and the dimming eye metaphor(00:00:57) - The retina stage: performers and crew(00:02:28) - Macular degeneration: overview and prevalence(00:04:19) - The genetic gap: regions vs genes(00:05:28) - Mendelian randomization narrows candidates(00:06:29) - Validating candidates across tissues(00:07:21) - Zebrafish screening of candidate genes(00:08:30) - Zebrafish results: CNN2 and SARM1 effects(00:09:38) - Focusing on CNN2 in mice(00:10:11) - Mouse vision tests: contrast and sensitivity(00:11:17) - Retinal structure changes in knockout mice(00:11:44) - Where CNN2 is expressed: vessels not photoreceptors(00:12:46) - What is CNN2 and proposed mechanism(00:13:14) - Caveats and limitations of the study(00:14:27) - Significance and future implications(00:15:05) - Summary and closing thoughts(00:16:08) - Credits and outro(00:16:49) - Original song: The dimming eye ballad

  2. há 6 dias

    461: Adult Ank3 loss quiets neurons and lowers a myelin protein

    Yoon et al., Proceedings of the National Academy of Sciences - ANK3, which encodes the scaffolding protein ankyrin-G, is a major risk gene for bipolar disorder and schizophrenia, yet what it does in adult neurons has been unclear. This study deletes Ank3 from mouse forebrain excitatory neurons either before birth or from adolescence and finds a convergent adult profile: hyperactivity and less anxiety-like and depression-like behavior, with social behavior intact. The neurons become less active, their membrane proteome shifts, and myelin basic protein falls, a protein made by oligodendrocytes rather than by the neurons that lost Ank3. Key terms: ANK3, ankyrin-G, bipolar disorder, myelin basic protein, lithium. Study Highlights: Deleting Ank3 prenatally with Emx1-Cre or from adolescence with CaMKIIα-Cre produced convergent adult phenotypes, including hyperactivity, reduced anxiety-like behavior and reduced immobility in the forced swim test, while the three-chamber social test was normal in both lines. Calcium imaging showed that ankyrin-G knockdown lowered spontaneous activity in cultured cortical neurons and that knockout slices gave smaller KCl-evoked calcium responses in layer 2/3 of somatosensory cortex. Quantitative TMT proteomics of the cortical membrane fraction quantified 2,357 proteins, of which 75 changed, 59 up and 16 down, and western blots confirmed higher Ryr2 and Taok2 and lower myelin basic protein. Mbp immunostaining was reduced in cortical layers 2/3 and 4 but not in layer 5 or hippocampal CA3, and myelin sheath diameter did not change significantly. A three-week lithium diet restored Mbp in the P3 fraction but not in the P2 membrane fraction, and left Ryr2 and Taok2 unchanged. Conclusion: Removing ankyrin-G from adult excitatory neurons is enough to produce hyperactive, low-anxiety mice with quieter neurons and less myelin basic protein, which ties a psychiatric risk gene in neurons to a myelin-associated protein made by glia. Lithium's selective restoration of Mbp suggests one possible myelin-related component of its action, although the cellular source of the Mbp change and any structural myelin deficit remain to be shown. Music: Enjoy the music based on this article at the end of the episode. Article title: Ank3 loss in adult forebrain excitatory neurons disrupts behavior, neuronal activity, membrane proteome, and myelination First author: Yoon Journal: Proceedings of the National Academy of Sciences DOI: 10.1073/pnas.2606900123 Reference: Yoon, S., Dos Santos, M., Khalatyan, N., Savas, J.N., and Penzes, P. (2026). Ank3 loss in adult forebrain excitatory neurons disrupts behavior, neuronal activity, membrane proteome, and myelination. Proceedings of the National Academy of Sciences 123(35), e2606900123. https://doi.org/10.1073/pnas.2606900123 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 ▶️ Watch with chapters on YouTube: https://www.youtube.com/@basebybase 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/ank3-adult-neurons-activity-myelin-basic-protein-lithium QC: This episode was checked against the original article PDF and publication metadata for the episode release published on 2026-09-24. QC Scope: - article metada... Chapters (00:00:20) - Introduction: The hyperactivity paradox(00:02:02) - Meet ANK3 and ankyrin G's structural role(00:03:52) - Mouse model design: timing the gene deletion(00:06:19) - Behavioral tests: hyperactivity and reduced anxiety(00:08:34) - Calcium imaging reveals quiet neurons(00:10:17) - Proteomics: widespread molecular remodeling(00:12:09) - The myelin basic protein surprise(00:13:57) - Localizing the MBP deficit in the cortex(00:14:36) - Lithium treatment and the P2 fraction results(00:16:10) - Lithium's precise rescue in the synaptic P3 fraction(00:17:08) - Hypothesis: axoglial signaling breakdown(00:18:28) - Rethinking how mood stabilizers work(00:19:17) - Limitations and study boundaries(00:20:06) - Summary and implications for future treatment(00:20:43) - Closing and credits(00:21:57) - Original inspired track

  3. 21 de set.

    460: The lupus variant that also sharpens antiviral defense

    Virolainen et al., The American Journal of Human Genetics - A lupus signal on chromosome 11p15 narrows to a coding haplotype in IRF7 that most people in the world carry. This study shows the risk form moves into the nucleus more readily, binds DNA more tightly and shifts its sequence preference, raising interferon-alpha output, and that mice engineered with the equivalent change clear a respiratory virus better while producing more anti-DNA autoantibodies. The same change buys antiviral protection and costs autoreactivity. Key terms: systemic lupus erythematosus, IRF7, type I interferon, evolutionary trade-off, fine mapping. Study Highlights: Fine mapping across ancestries placed the lupus association at 11p15 on a common coding IRF7 haplotype tagged by rs1131665 (p.Gln412Arg), with odds ratios of 1.20 in European and 1.15 in African ancestry cohorts and 1.18 in the trans-ancestral meta-analysis. The risk allele is carried by most of the world, reaching 98% in East Asian ancestry and 42% in African ancestry datasets, and ancient DNA spanning more than 40,000 years shows it was already common well before the recorded epidemics the authors examined. In cells the risk form increased IRF7 nuclear localization, strengthened DNA binding and altered its sequence specificity, raising interferon-alpha after stimulation of pattern recognition receptors in monocytes and airway epithelial cells. Mice carrying the analogous CRISPR-engineered Arg334Gln change in Irf7 had lower lung viral titers after intranasal vesicular stomatitis virus and produced significantly more anti-dsDNA autoantibodies than mice with the protective form. The authors read this as an evolutionary trade-off, in which the same gain in interferon output buys innate antiviral control at the price of autoreactivity. Conclusion: A coding haplotype in IRF7 that a majority of people carry tunes interferon-alpha upward, and that single setting produces both better early control of viral infection and a higher tendency to autoantibody production. Reading lupus risk at this locus as the cost of an antiviral advantage helps explain why such a variant stayed common, and it places interferon-directed therapy in a context where the same pathway is doing useful work. Music: Enjoy the music based on this article at the end of the episode. Article title: A highly prevalent lupus risk haplotype increases IRF7-dependent induction of IFN-α, enhancing antiviral defense and exacerbating autoimmunity First author: Virolainen Journal: The American Journal of Human Genetics DOI: 10.1016/j.ajhg.2026.08.016 Reference: Virolainen, S.J., Creighton, K., Dashtiahangar, M., et al., Waggoner, S.N., Weirauch, M.T., and Kottyan, L.C. (2026). A highly prevalent lupus risk haplotype increases IRF7-dependent induction of IFN-α, enhancing antiviral defense and exacerbating autoimmunity. The American Journal of Human Genetics 113, 1-22. https://doi.org/10.1016/j.ajhg.2026.08.016 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 ▶️ Watch with chapters on YouTube: https://www.youtube.com/@basebybase 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/irf7-lupus-haplotype-interferon-antiviral-tradeoff QC: This episode was checked against the o... Chapters (00:00:14) - Intro: The evolutionary paradox of a 'bad' gene(00:02:15) - What is lupus? Understanding the disease(00:03:17) - IRF7: The master regulator of interferon response(00:04:44) - The globally dominant lupus risk variant(00:05:59) - Tracing the variant through ancient DNA(00:07:33) - Engineering cells with CRISPR to test the variant(00:08:36) - Validating mouse models for human immunity(00:10:46) - Findings: ancient persistence and cellular mechanics(00:11:57) - Doubling interferon: the biological amplifier effect(00:14:19) - Mouse trials: autoimmunity vs viral survival tradeoff(00:15:31) - The evolutionary bargain: survive now, pay later(00:16:51) - Treatment implications and the risk of turning off IRF7(00:18:15) - Future research and remaining mysteries(00:19:11) - Summary and closing thoughts(00:20:14) - Closing credits and outro song(00:20:38) - Original track: 'Same Dial, Both Ways'

  4. 13 de set.

    459: Cerebral palsy genetics: 515 candidate genes, evidence for 89

    Arterbery et al., The American Journal of Human Genetics - Hundreds of genes have been reported as causes of cerebral palsy, yet there is no agreed model of what a pathogenic variant in a child with CP actually means. This study treats CP as a phenotypic feature that some genetic disorders make more likely, tests the reported genes against the population prevalence of CP across tens of thousands of published individuals, and finds statistical evidence of association for only 89 of 515, before applying that stratification to genome sequencing of 460 children with CP. Key terms: cerebral palsy, Mendelian disease, genome sequencing, Human Phenotype Ontology, gene-disease association. Study Highlights: Across 21 published genomic diagnostic cohorts, variants deemed causal of cerebral palsy had been reported in 515 genes, with no consensus on whether such variants cause CP, mimic it, or coexist with it. The authors modelled CP as a phenotypic feature with a known population prevalence and applied a Bonferroni-corrected binomial test to 280 genes drawn from 5,508 publications describing 43,897 individuals, rejecting the null hypothesis of no association for 86 genes, and for 89 when combined with the genes enriched across the CP sequencing cohorts. Genome sequencing of 460 children with CP from 453 families at Shriners Children's identified pathogenic or likely pathogenic variants in 60 genes in 15.8% of the children. Only 16 of those 60 genes had significant literature evidence for CP association, which still represented a 16.3-fold enrichment over the overlap expected by chance. The results support a stratified approach in which the strength of evidence for each gene shapes how a molecular finding in a child with CP is interpreted. Conclusion: A pathogenic variant found in a child with cerebral palsy does not carry the same weight for every gene: for most of the 515 genes reported so far, the literature does not yet provide statistical evidence that the associated disorder raises the likelihood of CP. Treating CP as a phenotype with graded genetic evidence offers a framework for attributing causality more carefully in precision genomic diagnosis, while leaving open that more genes may prove to be associated as data accumulate. Music: Enjoy the music based on this article at the end of the episode. Article title: A phenotypic paradigm for cerebral palsy genetics First author: Arterbery Journal: The American Journal of Human Genetics DOI: 10.1016/j.ajhg.2026.08.007 Reference: Arterbery, A.S., Gargano, M.A., et al., and Robinson, P.N. (2026). A phenotypic paradigm for cerebral palsy genetics. The American Journal of Human Genetics 113, 1972-1986. https://doi.org/10.1016/j.ajhg.2026.08.007 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 ▶️ Watch with chapters on YouTube: https://www.youtube.com/@basebybase 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/cerebral-palsy-genetics-phenotypic-model QC: This episode was checked against the original article PDF and publication metadata for the episode release published on 2026-09-13. QC Scope: - article metadata and core scientific claims from the narration - excludes analogies, intro/outro, and music...

  5. 10 de set.

    458: Somatic or inherited? Reading TP53 risk from shared DNA

    MacGregor et al., The American Journal of Human Genetics - Pathogenic TP53 variants found in blood have long been read as inherited Li-Fraumeni alleles, but many turn out to be somatic clones that grew with age. Using whole-exome data from 469,391 UK Biobank participants, this study combines variant allele fraction with haplotype sharing to tell the two origins apart, and finds that in cancer-free middle-aged adults the risk is mostly somatic and mostly hematological. Key terms: TP53, clonal hematopoiesis, Li-Fraumeni syndrome, UK Biobank, variant classification. Study Highlights: Rare coding TP53 variants were called in 469,391 UK Biobank participants aged 40 to 70, and 1,753 nonsynonymous variants were sorted by variant allele fraction combined with haplotype sharing across individuals carrying the same allele. Of 55 unique pathogenic TP53 variants detected, only six were classified as germline, and most of those sit at positions linked to partial rather than complete loss of p53 function. Classic Li-Fraumeni alleles behaved like somatic clonal expansions: their prevalence rose with age at an odds ratio of 2.3 per decade, while likely germline variants showed no age trend. Incident cancer risk separated sharply by origin, with the signal for somatic and inconclusive variants driven almost entirely by blood cancers rather than solid tumors, reaching a hazard ratio of 16.5 for the high-allele-fraction inconclusive class. The prevalence of somatic expansion tracked missense pathogenicity, suggesting that clonal growth in blood is itself an in vivo readout of how much a variant damages p53. Conclusion: In a population cohort of healthy middle-aged adults, most apparently pathogenic TP53 variants detected in blood reflect somatic clonal expansion rather than inherited predisposition, and the cancer risk they carry is predominantly hematological. Distinguishing the two origins with allele fraction and shared haplotypes gives a scalable way to interpret TP53 variants outside the clinic, where classic Li-Fraumeni assumptions do not hold. Music: Enjoy the music based on this article at the end of the episode. Article title: Shared inheritance reveals landscape of somatic and germline cancer risk in TP53 First author: MacGregor Journal: The American Journal of Human Genetics DOI: 10.1016/j.ajhg.2026.07.011 Reference: MacGregor, H.A.J., Blundell, J.R., and Easton, D.F. (2026). Shared inheritance reveals landscape of somatic and germline cancer risk in TP53. The American Journal of Human Genetics 113, 1916-1928. https://doi.org/10.1016/j.ajhg.2026.07.011 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 ▶️ Watch with chapters on YouTube: https://www.youtube.com/@basebybase 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/tp53-somatic-germline-cancer-risk-biobank QC: This episode was checked against the original article PDF and publication metadata for the episode release published on 2026-09-10. QC Scope: - article metadata and core scientific claims from the narration - excludes analogies, intro/outro, and music - transcript coverage: Audited the scientific body from the TP53 and Li-Fraumeni background through clonal hematopoiesis, the allele-f...

  6. 9 de set.

    457: A deletion that raises Alzheimer risk, a duplication that lowers it

    Quenez O et al., The American Journal of Human Genetics - Rare copy-number variants were called from 22,319 exomes covering early-onset Alzheimer disease, late-onset disease and unaffected controls, then tested gene by gene for a dosage effect. One locus came back with the cleanest signal in the field: at the central 22q11.21 region, deletions appeared only in early-onset cases, including one that arose de novo, while duplications piled up in controls, with late-onset cases sitting in between. Replication in nearly 400,000 further individuals confirmed it, and overexpressing SCARF2, one of the genes in the narrowed interval, increased amyloid-beta uptake in cells - the direction the protective duplications would predict. Key terms: copy-number variants, Alzheimer disease, 22q11.21, SCARF2, gene dosage. Study Highlights: The authors called rare copy-number variants from 22,319 harmonized exomes, comprising 4,150 early-onset Alzheimer cases at 65 years or younger, 8,519 late-onset cases and 9,650 unaffected controls, and found that early-onset cases carried an excess burden of rare CNVs affecting coding genes, driven particularly by deletions in genes already implicated in the disease. Pooling short truncating variants with deletions into a single loss-of-function analysis implicated ABCA1 with an odds ratio of 5.77 and ABCA7 deletions with an odds ratio of 2.29, and raised CTSB as a new candidate at an odds ratio of 5.03. An exome-wide gene-level dosage analysis then flagged 18 genes across five loci below a 10 percent false discovery rate, of which the central 22q11.21 region stood out: deletions were restricted to early-onset cases and absent from all 9,559 controls, duplications were enriched in controls, and late-onset cases fell in between, a gradient the authors narrowed to the SCARF2-KLHL22-MED15 interval. Replication across 33,977 additional cases and 362,322 controls confirmed the association at exome-wide significance, with duplications carrying an odds ratio of 0.34 and a mega-analysis p value of 5.52 times ten to the minus seven. Transfecting SCARF2 into microglial cells significantly increased uptake of fluorescent amyloid-beta with a p value of 2.2 times ten to the minus five, while uptake of inert latex beads did not change, arguing the effect is specific rather than a general boost in cell appetite. Conclusion: The same stretch of chromosome 22 pushes Alzheimer risk in opposite directions depending on how many copies you carry: losing one raises risk, gaining one lowers it. Duplications of the central 22q11.21 region emerge as one of the strongest risk-decreasing genetic factors described so far in terms of effect size, behind only the ultra-rare Icelandic APP variant, and unlike that variant they are found across multiple ancestries. The authors are careful about what this does not mean: the deletions are extremely rare, protection is relative rather than absolute, and they show it directly by pointing to duplication carriers who developed late-onset disease anyway while carrying other risk factors. Power is the central constraint, because each genome carries hundreds of rare coding point variants but typically only one rare coding CNV, and the analysis was restricted to individuals of European ancestry. Music: Enjoy the music based on this article at the end of the episode. Article title: Exome analysis of 22,319 individuals links extremely rare copy-number variants and 22q11.21 dosage to Alzheimer risk First author: Quenez O Journal: The American Journal of Human Genetics DOI: 10.1016/j.ajhg.2026.07.013 Reference: Quenez O, Schramm C, Cassinari K, Nicolas A, Groeneveld J, Huguet G, et al. Exome analysis of 22,319 individuals links extremely rare copy-number variants and 22q11.21 dosage to Alzheimer risk. The American Journal of Human Genetics. 2026;113(10):1-19. doi:10.1016/j.ajhg.2026.07.013 Li...

  7. 8 de set.

    456: Beyond exons: where heritability hides as traits get more polygenic

    Fuhrer J et al., The American Journal of Human Genetics - Across 34 complex traits and disorders, a MiXeR-based framework partitions SNP heritability over 74 functional annotations and finds that exons carry only a minority of it, and steadily less as a trait becomes more polygenic. Exonic heritability falls from about 22 percent in less-polygenic somatic diseases and biomarkers to about 13 percent in highly polygenic psychiatric and cognitive traits, intergenic heritability rises by the same logic, and intronic heritability stays put. A new annotation contribution score shows the same axis in the annotations themselves: highly polygenic traits load on conservation and variant-effect scores, less-polygenic traits on promoter, transcription and chromatin marks. Key terms: polygenicity, SNP heritability, noncoding variation, functional annotation, complex traits. Study Highlights: The authors extended the MiXeR framework to model SNP effect-size variance across 74 binary functional annotations, applied it to 34 traits spanning psychiatric, neurological, cardiometabolic, anthropometric, hematological and immune phenotypes, and introduced a likelihood-based annotation contribution score that measures how much a single annotation adds to the full model rather than how enriched it is per SNP. Exons cover 2.55 percent of base pairs and account for a mean of 14.52 percent of heritability, with introns and intergenic regions together explaining the remaining 85.49 percent; introns alone typically carry about half. Regressing regional fractions on log polygenicity, the exonic fraction falls 4.38 percentage points and the intergenic fraction rises 4.87 percentage points per ten-fold increase in polygenicity, while the intronic slope is not significant, and the illustrative spread runs from 8.51 percent exonic in schizophrenia to 29.44 percent in sex-hormone-binding globulin. The contribution score peaks for mid-sized annotations covering roughly 10 to 15 percent of SNPs, so compact conservation tracks such as phastCons, GERP, CADD and Eigen dominate in highly polygenic traits while promoter and chromatin tracks such as H3K4me3 and CpG islands dominate in less-polygenic ones. Sensitivity analyses against sLDSC and sLD4M agreed closely, with heritability fractions correlating at r equals 0.996 and polygenicity estimates at Spearman r equals 0.91. Conclusion: The functional location of heritability is not a fixed property of the genome but tracks a trait's polygenicity: less-polygenic somatic traits concentrate heritability in coding and gene-proximal regulatory regions, while highly polygenic psychiatric and cognitive traits distribute it across dispersed distal regulatory elements. Because enrichment per SNP and total contribution can point in opposite directions, the authors argue both must be read together, and they note that coding variants explain a minority of heritability throughout, which argues for sequencing designs that maximize noncoding coverage. The analysis relies on European-ancestry reference panels, uses relatively broad annotations with incomplete chromosome X coverage, and depends on polygenicity estimates from two methods that nonetheless agreed closely. Music: Enjoy the music based on this article at the end of the episode. Article title: Beyond exons: Linking noncoding heritability and polygenicity across complex human traits and disorders First author: Fuhrer J Journal: The American Journal of Human Genetics DOI: 10.1016/j.ajhg.2026.08.012 Reference: Fuhrer J, Shadrin AA, Hughes T, Parker N, Hindley G, Frei E, Nguyen D, Smeland OB, Djurovic S, Andreassen OA, Dale AM, Frei O. Beyond exons: Linking noncoding heritability and polygenicity across complex human traits and disorders. The American Journal of Human Genetics. 2026;113(10):1-13. doi:10.1016/j.ajhg.2026.08.012 License: This episode is based...

  8. 7 de set.

    455: Agentic genomics: the bottleneck moves from code to judgment

    Corpas M et al., Cell Genomics - A Perspective arguing that autonomous AI agents which discover, configure and chain bioinformatics operations from natural-language instructions have shifted the bottleneck in computational biology from building pipelines to validating their output. The authors define four necessary conditions for a system to count as agentic, propose a perturbation test that separates genuine runtime decision-making from pre-specified branching, survey four existing systems, and set out a three-tier validation framework in which none of the surveyed systems yet reaches clinical grade. Key terms: agentic genomics, AI agents, validation, skill libraries, equity in genomics. Study Highlights: The authors define agentic genomics through four necessary conditions — autonomy, domain constraint via validated skill libraries, iterative refinement, and natural-language mediation — and offer an empirically testable criterion: a system is not agentic if perturbed intermediate outputs fail to change its execution strategy. They distinguish the paradigm from workflow managers like Nextflow and Galaxy, from AutoML, from LLM-assisted scripting, and from general-purpose biomedical copilots. Four systems are surveyed and placed against a proposed three-tier validation framework of research grade, benchmarked and clinical grade; all four currently sit at research grade, and none has published the external multi-site evidence the clinical tier requires. The paper documents a concrete silent failure: in the early weeks of ClawBio, one skill given an empty input file containing no genomic data returned all-normal results including recommended dosages for 51 drugs, a path since closed by community audit. It also reports that a randomized trial of GPT-4-augmented diagnostic reasoning found no significant improvement over conventional resources, 76 percent versus 74 percent with p equal to 0.60, even though the model alone scored well above physicians. Conclusion: The central claim is that agentic genomics expands the capacity to produce analyses without expanding the capacity to judge them, so validation frameworks must be formalized and tiered, equity-aware defaults must be engineered rather than aspired to, and regulatory frameworks must accommodate non-deterministic software. The authors argue the paradigm is already in use and advancing, and that the open question is not whether it will be adopted but whether the field will build the standards to make it trustworthy first. Music: Enjoy the music based on this article at the end of the episode. Article title: Agentic genomics: From pipeline automation to autonomous validation First author: Corpas M Journal: Cell Genomics DOI: 10.1016/j.xgen.2026.101305 Reference: Corpas M, Guio H, Fatumo S. Agentic genomics: From pipeline automation to autonomous validation. Cell Genomics. 2026;6(8):101305. doi:10.1016/j.xgen.2026.101305 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 ▶️ Watch with chapters on YouTube: https://www.youtube.com/@basebybase 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/agentic-genomics-validation-bottleneck QC: This episode was checked against the original article PDF and publication metadat...

Ver tudo (463)

Sobre

Base by Base explores advances in genetics and genomics, with a focus on gene-disease associations, variant interpretation, protein structure, and insights from exome and genome sequencing. Each episode breaks down key studies and their clinical relevance—one base at a time. Powered by AI, Base by Base offers a new way to learn on the go. Special thanks to authors who publish under CC BY 4.0, making open-access science faster to share and easier to explore.

Informações

  • Criado por
    Gustavo Barra
  • Anos de atividade
    2025 - 2026
  • Episódios
    463
  • Classificação
    Livre
  • Copyright
    © Gustavo Barra
  • Site do podcast
    Base by Base