Manoj K Valluru (@mkvalluru) header
Manoj K Valluru (@mkvalluru) profile picture

https://nitter.cf/t.co/7Xqee6zmyZ Postdoc @Ex_vivo_group @ShefUni_ClinMed

Yorkshire and The Humber
Joined August 2010
Photo by Manoj K Valluru (@mkvalluru)Photo by Manoj K Valluru (@mkvalluru)Photo by Manoj K Valluru (@mkvalluru)Photo by Manoj K Valluru (@mkvalluru)Photo by Manoj K Valluru (@mkvalluru)Photo by Manoj K Valluru (@mkvalluru)Photo by Manoj K Valluru (@mkvalluru)Photo by Manoj K Valluru (@mkvalluru)Photo by Manoj K Valluru (@mkvalluru)Photo by Manoj K Valluru (@mkvalluru)
Manoj K Valluru retweeted
A pervasive RT–qPCR artifact inflates RNA knockdown by RNA-targeting CRISPR - Yikey Yikes! BTW, similar concerns apply to siRNAs/ASOs that are sometimes suspected to affect RT and/or PCR . So nice somebody looked into this! rdcu.be/RUYFaGa649va
11
1
60
8,441
Manoj K Valluru retweeted
I am so pleased to see this paper by Andy Streets et al @shefunihealth in press. Well done. An integrin beta-1-anchored bicaudal C1-polycystin-1 module essential ... sciencedirect.com/science/ar…
2
1
116
“An integrin beta-1-anchored bicaudal C1–polycystin-1 module essential …” sciencedirect.com/science/ar… A big thank you to all my co-authors for the collaboration, contributions and hard work that made this possible. @Ong_Lab @ChrisToseland @Kidney_Research
1
3
767
Manoj K Valluru retweeted
Oxygen-dependent and photoreactive APEX2-based proximity labeling using a chemically evolved biotin probe for applications in living cells and in vivo nature.com/articles/s41589-0…
12
1
64
5,989
Manoj K Valluru retweeted
Essential lncRNAs in the human transcriptome @CellGenomics @nevillesanjana @nyuniversity cell.com/cell-genomics/fullt…
11
42
3,111
New to OME-Zarr? Join us April 15, 🕒️ 15:00–17:30 CEST for a guided tour of this cloud‑native format for large bioimaging data. Learn conversion, visualisation, validation, & metadata editing using EuBI-Bridge. #OMEZarr #BioData #FAIRdata #DataFormats embl-org.zoom.us/meeting/reg…
4
4
305
Manoj K Valluru retweeted
Compressing the collective knowledge of ESM into a single protein language model @naturemethods 1. The paper argues that “sequence-only” protein language models (PLMs) are not intrinsically capped for variant-effect prediction (VEP); instead, their evolutionary signals are fragmented across model families and can be recovered by making models learn from each other. 2. Key observation: closely related ESM models have complementary blind spots. For example, ESM2 models systematically miss KRAB-domain conservation signals, while ESM1b/ESM1v can miss BRICHOS-domain signals; yet at least one model in the family captures each domain’s mutational sensitivity. 3. They introduce a simple but effective ensemble rule: for each missense mutation, take the minimum log-likelihood ratio (LLR) across models (ESMIN), i.e., “maximum confidence” scoring. This can amplify subtle evolutionary constraints that averaging would dilute. 4. A theoretical analysis explains when min-LLR beats averaging: if pathogenic-variant LLRs are more dispersed across models than benign-variant LLRs (variance asymmetry). The ESM family empirically shows this property, making maximum-confidence aggregation advantageous. 5. ESMIN is evaluated using 11 sequence-only ESM models (ESM1b, five ESM1v, five ESM2; excluding ESM2-15B). It outperforms averaging-based ensembles and improves ProteinGym DMS correlations, with gains occurring in ~50% of assays (versus ~20% for typical ensembles). 6. Main methodological contribution: “maximum-confidence co-distillation.” For each protein, all models score all mutations; the element-wise minimum LLR matrix becomes a teacher signal, and each model is trained (variant-level MSE) to match these confident targets—without MSAs, structures, or population genetics features. 7. Co-distillation substantially improves every participating model, including small ones: ESM2-8M improves on ClinVar AUC from ~0.65 to ~0.88. Several co-distilled single models (e.g., ESM2-3B, ESM1b, ESM2-650M) can even surpass the ESMIN teacher signal (“student surpasses teacher”). 8. Robustness/ablation: improvements persist when training data are heavily reduced and de-homologized. With only ~1% of human proteins (~200 sequences; <30% identity to benchmark proteins), ESM2-35M reaches ~97% (ClinVar) and ~94% (DMS) of its peak co-distilled performance. 9. Iterative procedure: after round 1 (min-LLR co-distillation), additional rounds switch to average-aggregation co-distillation. As models improve, class-conditional variances become more symmetric, making averaging slightly better; after 3 rounds, a single 3B model matches the ensemble—named VESM-3B. 10. Practical compression: VESM-3B is distilled into smaller models (650M, 150M, 35M) that retain most performance (reported as >98% on Balanced ClinVar and >93% on ProteinGym DMS relative to VESM-3B), enabling high-throughput VEP under limited compute. 11. Clinical benchmark (ProteinGym ClinVar, 2,227 genes): sequence-only VESM models outperform other sequence-only PLMs (including ESM-C) and compete with or surpass methods using MSA/structure/population priors. VESM-3B shows balanced ROC behavior across specificity and sensitivity regimes. 12. AlphaMissense comparison: VESM-3B performance is stable across allele-frequency strata, while AlphaMissense shows strong dependence on MAF (consistent with circularity risks when population frequency informs clinical labels). After excluding variants overlapping AlphaMissense training (gnomAD v2 MAF > 1e-5), all VESM sizes outperform AlphaMissense on AUC and multiple calibrated metrics. 13. Modular use of structure: rather than retraining a joint model, they fine-tune the sequence component of ESM3 using VESM-style sequence-based loss to create VESM3, and combine VESM3 with VESM-3B into a structure-aware ensemble (VESM++). This improves performance on structure-dependent DMS assays (binding/stability/expression) while maintaining strong fitness/activity performance. 14. Cross-domain generalization: despite co-distillation being trained on human proteins, gains transfer strongly to nonhuman DMS assays, with disproportionately large improvements reported for viral proteins—even though ESM3’s released training data excluded viral sequences. 15. Beyond binary pathogenicity: using UK Biobank/Genebass summary statistics for 332 gene–phenotype pairs (blood biochemistry biomarkers), variant-level VESM scores correlate with single-variant effect sizes (β). VESM++ and VESM-3B yield the strongest gene–trait association signals across tested models. 16. Notably, VESM-3B recovers the correct pLoF direction of effect in 98.8% of significant gene–phenotype pairs and identifies many associations not detected by missense burden tests, suggesting utility for quantitative trait interpretation from summary statistics. 📜Paper: doi.org/10.1038/s41592-026-0… #ProteinLanguageModels #VariantEffectPrediction #ComputationalBiology #HumanGenetics #ESM #ClinVar #ProteinGym #DeepMutationalScanning #UKBiobank #MachineLearning
1
27
2
123
22,304
Manoj K Valluru retweeted
We are looking for an outstanding postdoc to join my lab at the Human Technopole in Milan, to tackle big questions in the biology and function of long non-coding RNAs. Details here: careers.humantechnopole.it/j…
1
24
1
58
7,385
"The AI Scientist, writes code, runs experiments, plots and analyses data, writes the entire scientific manuscript, and performs its own peer review." Amazingly, this is not satire! It's from a new paper in @Nature. Maybe they meant to release it on April 1? 🤣
1
2
16
5,083
Manoj K Valluru retweeted
🐍 Learn Python for Image Analysis! UK life scientists: join EuroBioImaging's hands-on beginner course at The Crick, London (23–25 June), running with sites in CZ, PT and SWE! ⏰ Register by 15/4— places are limited! ✈️ Travel funding may be available shorturl.at/v3Vsk
18
2
43
3,564
New preprint on how disagreement among variant effect predictors (VEPs) can help guide prioritization of proteins for experimental analysis We analyse for which proteins VEPs disagree, what features they have, and suggest lack of concordance & clinical data to guide experiments
Disagreement among variant effect predictors guides experimental prioritization of target proteins biorxiv.org/content/10.64898… #biorxiv_bioinfo
3
18
92
10,792
TLにも流れていた、マウス全細胞(胎児は全身、成体は全臓器)の3次元アトラス論文。今週のCELL誌の表紙をとった。 cell.com/cell/abstract/S0092…
1
7
74
5,862
Manoj K Valluru retweeted
We are recruiting! If you are passionate about technology development, protein engineering, computational design, directed evolution, chemical biology - please reach out! (The setting is pretty nice too…)
23
70
5
508
61,663
Manoj K Valluru retweeted
The Human Organ Atlas, a new resource for researchers, clinicians, and educators, is an open-access database of 3D imaging of intact human organs. The portal includes donor samples with conditions from congenital disorders to COVID-19. Learn more in @ScienceAdvances: scim.ag/4bnSEzZ
1
62
4
159
20,883
Manoj K Valluru retweeted
Mechanical forces play essential roles in biology. Measuring and controlling these forces represent emerging strategies for disease diagnosis and therapeutic intervention, to complement more traditional approaches. This review article, published today in Nature Reviews Bioengineering (nature.com/articles/s44222-0…), summarizes the status of work in this area -- from organisms, organs and tissues to cells and molecules -- and future directions. Check it out if you’d like to learn more about this field of research and translation. Big thanks to my senior co-authors, Prof. @songli_UCLA, Prof. Jun Chen and Prof. Viola Vogel, for their leadership on this piece, to Dr. Min-Seung Jo (postdoc in the group) for his contributions to our sections on sensors and actuators, and to the other co-authors for their roles.
2
12
65
5,076
Dear students in the UK, if you haven't had your teenage meningitis vaccine or aren't sure if you have had it, PLEASE arrange with your GP to get vaccinated nhs.uk/vaccinations/menacwy-… It helps prevent tragic deaths like those that have occured recently in Kent.
48
648
19
990
108,768
Manoj K Valluru retweeted
We’ve trained a multimodal AI model to turn routine pathology slides into spatial proteomics, with the potential to reduce time and cost while expanding access to cancer care.
444
1,784
369
10,897
2,812,780
Manoj K Valluru retweeted
DNAsight: an automated analysis framework for AFM data that integrates machine learning (ML)-based segmentation with modular, base-pair-calibrated quantification of DNA spatial organization, looping, nucleosome spacing, and protein clustering biorxiv.org/content/10.64898…
1
6
512