Sign in

Sonika Tyagi

@tsonika.bsky.social
150 followers 258 following 96 posts

Academic @RMIT University Melbourne Australia #computationalbiology #digitalhealth #MachineLearning #AI #LLM #genomics #data4diversity #aiforsocialgood #equityDiversityInclusion

PostsRepliesMedia
Reposted by Sonika Tyagi
AI in Digital Health & Bioinformatics Lab @tyagilab.bsky.social · 04/09/2026
lnkd.in/p/gTtrqTku PregBase is published in 𝗔𝗿𝘁𝗶𝗳𝗶𝗰𝗶𝗮𝗹 𝗜𝗻𝘁𝗲𝗹𝗹𝗶𝗴𝗲𝗻𝗰𝗲 𝗶𝗻 𝗠𝗲𝗱𝗶𝗰𝗶𝗻𝗲. The first comprehensive knowledge base designed specifically for pregnancy research. Try it: pregknowledgebase.com Read the full paper: lnkd.in/gs7W9KdS #PregBase #MaternalHealth #PhD #RMIT #DigitalHealth #TyagiLab
lnkd.in
New research out of STEM College has developed PregBase - the first knowledge base built specifically to bring together decades of pregnancy research in one place. PregBase is an AI system that… | RM...
New research out of STEM College has developed PregBase - the first knowledge base built specifically to bring together decades of pregnancy research in one place. PregBase is an AI system that reads...
212
Sonika Tyagi @tsonika.bsky.social · 31/07/2026
We are hosting the 2026 @abacbs.bsky.social conference at RMIT University City Campus Melbourne on 16-20 Nov 2026. Hope to see you there.
274
Sonika Tyagi @tsonika.bsky.social · 08/07/2026
New paper from the lab on pre-prints now. Led by PhD student @muraliaadhitya.bsky.social
022
Sonika Tyagi @tsonika.bsky.social · 13/05/2026
well done team!
000
Sonika Tyagi @tsonika.bsky.social · 10/05/2026
www.tandfonline.com/doi/full/10....
tandfonline.com
Increasing Data Literacy in High School Students Through Data Science and Healthcare: Part 1 of 2 Leading to AI Instruction
High school science and biomedical pathway teachers need effective strategies to build student data literacy and prepare them to conduct experiments. This first article of two presents the data col...
000
Sonika Tyagi @tsonika.bsky.social · 06/03/2026
🔊Applications invited: hashtag#Funded BITS–RMIT hashtag#PhD program offering international experience. We have two #WomenHealth oriented #DigitalHealth projects listed. These projects are designed to look at unique Australian and Indian context of different womens health outcome questions.
lnkd.in
LinkedIn
This link will take you to a page that’s not on LinkedIn
124
Reposted by Sonika Tyagi
abacbs.bsky.social @abacbs.bsky.social · 02/03/2026
We are delighted to announce that #ABACBS2026 will be held at RMIT, Melbourne CBD campus from 16th to 20th November 2026. Big thanks to A/Prof. Sonika Tyagi and team for leading this years organising! More details and official conference launch to follow! #bioinformatics @combine1.bsky.social
0106
Sonika Tyagi @tsonika.bsky.social · 23/02/2026
Kudos to Sam and @aashishbhandari.com !
010
Sonika Tyagi @tsonika.bsky.social · 30/01/2026
#MultiOmics data #integration using #MachineLearning. A book chapter co-authored with PhD student Samitha Amarapathy.
020
Sonika Tyagi @tsonika.bsky.social · 22/01/2026
econpapers.repec.org/article/plop...
econpapers.repec.org
EconPapers: Moving beyond the empty cell: The threat of decontextualized healthcare data
By Aya El Mir, Eric Bezerra de Sousa, Ignacio Mesina-Estarrón, Leo Anthony Celi, Moad Hani, Mohammed Benjelloun, Neha Nageswaran, Saïd Mahmoudi, Shaheen Siddiqui, Sreeram Sadasivam and William Greig Mitchell; Abstract: Missing, inaccurate, or poorly documented data in healthcare is often treated as a technical problem to be
010
Sonika Tyagi @tsonika.bsky.social · 21/01/2026
Deadline extended to 3 Feb 2026 #Spreadtheword
010
Sonika Tyagi @tsonika.bsky.social · 21/01/2026
Lead author @aashishbhandari.com
030
Sonika Tyagi @tsonika.bsky.social · 21/01/2026
‼️ New Publicaiton alert ‼️ Delighted to share the publication of our article “A comparative evaluation of handling missing data points and modalities in electronic health records” accepted in the International Journal of Medical Informatics. www.sciencedirect.com/science/arti...
sciencedirect.com
A comparative evaluation of handling missing data points and modalities in electronic health records
Background: Healthcare data, generally available as electronic health records (EHR), provide rich insight for predictive modelling. A common challenge…
120
Sonika Tyagi @tsonika.bsky.social · 13/01/2026
www.nature.com/articles/s41...
nature.com
Client Challenge
100
Sonika Tyagi @tsonika.bsky.social · 13/01/2026
#repost in your network. Postdoctoral Fellowship with @tyagilab.bsky.social and Barcelona Supercomputing Centre.
010
Sonika Tyagi @tsonika.bsky.social · 19/12/2025
Thank you 🙏🏼
000
Sonika Tyagi @tsonika.bsky.social · 18/12/2025
Conference diary Dec 2025
010
Sonika Tyagi @tsonika.bsky.social · 20/11/2025
Team day out @tyagilab.bsky.social
021
Reposted by Sonika Tyagi
AI in Digital Health & Bioinformatics Lab @tyagilab.bsky.social · 21/10/2025
Our lab head, @tsonika.bsky.social, presented at the 2025 Victorian Affiliation of Research Nurses Inc. Conference on the applications of AI in clinical trials. The session sparked a great exchange of ideas with delegates, highlighting the excitement and opportunities in this rapidly evolving field.
011
Sonika Tyagi @tsonika.bsky.social · 15/10/2025
Preprint from the lab.
010
Sonika Tyagi @tsonika.bsky.social · 15/10/2025
@aashishbhandari.com
010
Sonika Tyagi @tsonika.bsky.social · 14/10/2025
Our invited editorial in BMC Artificial Intelligence along with @naimavahab.bsky.social
111
Sonika Tyagi @tsonika.bsky.social · 14/10/2025
New publication from the lab
011
Reposted by Sonika Tyagi
Scott C Edmunds @scedmunds.bsky.social · 14/10/2025
Nice there’s another ML standards case study in this #IDW2025 “long-tail of supplemental data” session, Slava Tykhonov presenting on Croissant ML docs.google.com/presentation...
docs.google.com
croissant-deem-slides
Croissant: A Metadata Format for ML-Ready Datasets June 9, 2024 8th Workshop on Data Management for End-to-End Machine Learning (DEEM) 2024 SIGMOD/PODS 2024 Mubashara Akhtar, Omar Benjelloun, Costanza...
021
Sonika Tyagi @tsonika.bsky.social · 13/10/2025
Our Digital Health and AI applications reading group is on this week. Join in.
020
Reposted by Sonika Tyagi
Damiano Spina @damianospina.com · 10/10/2025
The pre-print of our #SIGIRAP2025 paper titled "ISMIE: A Framework to Characterize Information Seeking in Modern Information Environments" is now available arxiv.org/abs/2510.07644 @shuoqisun.bsky.social @danulah.bsky.social @admscentre.org.au @rmitcomputing.bsky.social
arxiv.org
ISMIE: A Framework to Characterize Information Seeking in Modern Information Environments
The modern information environment (MIE) is increasingly complex, shaped by a wide range of techniques designed to satisfy users' information needs. Information seeking (IS) models are effective mecha...
043
Reposted by Sonika Tyagi
Albert Vilella, PhD. @albertvilella.bsky.social · 08/10/2025
@ewanbirney.bsky.social on the new version of Alphafold at the @emblebi.bsky.social www.ebi.ac.uk/about/news/t... AlphaFold has been a transformation not just in structural biology but also - due to its ability to be applied to all proteins - to molecular biology as a whole.
ebi.ac.uk
EMBL-EBI and Google DeepMind renew partnership and release update to AlphaFold Database
Deeper collaboration between EMBL-EBI and Google DeepMind brings updates to the AlphaFold Database.
1207
Reposted by Sonika Tyagi
Stephen Turner @stephenturner.us · 08/10/2025
My colleagues and I here at UVA are looking to hire a skilled #Rstats developer. You'll be working with me on the project. Here's the posting: jobs.virginia.edu/us/en/job/R0...
jobs.virginia.edu
Epidemiologist / Research Data Scientist; Infectious Diseases and International Health in Charlottesville, Virginia, United States of America | Information Management, Services & Technology at Univers...
Apply for Epidemiologist / Research Data Scientist; Infectious Diseases and International Health job with University of Virginia in Charlottesville, Virginia, United States of America. Information Man...
039
Reposted by Sonika Tyagi
ICT&health @icthealth.nl · 08/10/2025
ChatGPT scores up to 97% on medical facts — yet fails to read symptoms. Experts like Ahmed Abdeen Hamed and Dr Andrew Thornton warn: AI can inform, not diagnose. The real challenge now? Using it responsibly in healthcare. #healthcare
icthealth.org
AI performs well on facts, but fails when it comes to symptoms | ICT&health Global
An important explanation lies in the nature of the language ChatGPT uses. Even more worrying is that it does not indicate when it does not know something.
021
Reposted by Sonika Tyagi
ElleM @ellem111.bsky.social · 03/10/2025
"Your best defense isn’t your eyes — it’s your brain. Keep practicing media and digital literacy..." #discover #MedSky #MentalHealth #AI www.medscape.com/viewarticle/...
medscape.com
Deepfake Doctors: How AI Spreads Medical Disinformation
AI-generated doctors, some fabricated and some mimicking well-known physicians, sell sketchy products and bad medical advice on social media. Here’s how it works.
044
Reposted by Sonika Tyagi
Open Data Science Conference @odsc.bsky.social · 02/10/2025
Discover how AI can help creators beat creative block, visualize ideas, and craft standout portfolios—without replacing human creativity. #DataScience #AI #ArtificialIntelligence opendatascience.com/the-ai-advan...
opendatascience.com
The AI Advantage: How Creators Can Use AI to Generate Portfolio Concepts
The creative industry is rapidly evolving, and a new co-pilot has entered the studio: artificial intelligence. Creators across fields—graphic designers, photographers, illustrators, writers—often face...
021
Reposted by Sonika Tyagi
Professor Euan Ritchie FAMS @euanritchie.bsky.social · 02/10/2025
A solemn day as we pay our respects to Dr Jane Goodall and her extraordinary legacy. Vale, Jane, we promise to keep fighting for all that you loved and fought for too! theconversation.com/only-if-we-h... With @drkyliesoanes.bsky.social, Marissa Parrott, @drvanessapirotta.bsky.social & Zara Bending.
theconversation.com
‘Only if we help shall all be saved’: Jane Goodall showed we can all be part of the solution
Jane Goodall showed tremendous courage in charting her own course as a pioneering researcher – and working to spread hope wherever she went.
314044
Reposted by Sonika Tyagi
naimavahab.bsky.social @naimavahab.bsky.social · 25/09/2025
A versatile machine learning pipeline to find co-regulatory modules (CRMs) in DNA. Check it out and explore: doi.org/10.1016/j.co... #Genomics #Epigenomics #MachineLearning #CardiacResearch #OpenScience @tyagilab.bsky.social @tsonika.bsky.social
041
Reposted by Sonika Tyagi
Aashish Bhandari @aashishbhandari.com · 25/09/2025
🩺 Missing medical data isn't just something to fill in or ignore! EHRs often have missing values. Common fix? Imputation. But filling gaps can mislead predictions. We explore ML approaches to handle missingness while preserving the original data distribution. 📜 www.researchsquare.com/article/rs-6...
researchsquare.com
Mind the Gaps: Guess Less, Predict More with Missing Medical Data
Healthcare data, generally available as electronic health records (EHR), provide a rich profile of an individual’s health and lifestyle. This data can be harnessed for predictive modelling using machi...
012
Sonika Tyagi @tsonika.bsky.social · 24/09/2025
Congratulations to @yashpalr.bsky.social on his graduation! @tyagilab.bsky.social
141
Reposted by Sonika Tyagi
Professor Asha Rao @asharao.bsky.social · 20/09/2025
Sharenting is the name given to the action of parents sharing photos of their children online. www.theguardian.com/technology/2...
theguardian.com
Parents outraged as Meta uses photos of schoolgirls in ads targeting man
Exclusive: Instagram pictures of girls as young as 13 were posted to promote Threads site ‘as bait’, campaigner says
031
Reposted by Sonika Tyagi
Claire Foldi @foldi.bsky.social · 07/09/2025
Incredibly proud of @kxhuang.bsky.social for persisting with such a challenging dataset from her PhD! We found some intriguing genetic targets in a corticostriatal circuit that may be involved in the development of #anorexia nervosa 🤔 onlinelibrary.wiley.com/doi/10.1002/...
onlinelibrary.wiley.com
Toward a Genetic Signature of Resistance to Activity‐Based Anorexia in Striatal Projecting Cortical Neurons
Objective Converging evidence from neuroimaging studies and genome-wide association study (GWAS) suggests the involvement of prefrontal cortex (PFC) and striatum dysfunction in the pathophysiology o...
3104
Reposted by Sonika Tyagi
Lorne Genome Conference, Australia @lornegenome.bsky.social · 03/09/2025
Earlybird registration & abstract submissions now open for the 𝟒𝟕𝐭𝐡 𝐋𝐨𝐫𝐧𝐞 𝐆𝐞𝐧𝐨𝐦𝐞 𝐂𝐨𝐧𝐟𝐞𝐫𝐞𝐧𝐜𝐞, 𝟏𝟓-𝟏𝟖𝐭𝐡 𝐅𝐞𝐛 𝟐𝟎𝟐𝟔: www.lornegenome.org/registrations (deadline 7th Nov 2025) Speakers: www.lornegenome.org/speakers We look forward to seeing you there!
lornegenome.org
Speakers — Lorne Genome
No description found
064
Sonika Tyagi @tsonika.bsky.social · 04/09/2025
Kudos to Naima Vahab for presenting her PhD research at "Storytelling for Impactful Research (STIR) Competition " at RMIT STEM College. What an amazing opportunity for polishing your #scicomm skills and knowledge sharing across STEM disciplines. @tyagilab.bsky.social @rmitcomputing.bsky.social
222
Reposted by Sonika Tyagi
Bioinformatics Advances @bioinfoadv.bsky.social · 13/06/2025
🧠 Just out in Bioinformatics Advances: “Predicting gene expression using millions of yeast promoters reveals cis-regulatory logic”   Explore the paper here: doi.org/10.1093/bioadv/vbaf130 Authors include: @susbo.bsky.social
142
Reposted by Sonika Tyagi
tmel1219.bsky.social @tmel1219.bsky.social · 13/08/2025
#ProudBlue #Voices4Victory Science Matters! Scientists have found protein fragments from the Covid-19 virus hidden inside tiny cellular packages in the blood of long-covid patients, offering the first potential measurable biomarker for the condition. www.sciencedaily.com/releases/202...
sciencedaily.com
Scientists detect virus traces in blood that may unlock long COVID’s mystery
Scientists have found protein fragments from the COVID-19 virus hidden inside tiny cellular packages in the blood of long COVID patients, offering the first potential measurable biomarker for the cond...
610266
Reposted by Sonika Tyagi
David Lingenfelter, PhD @dlingenfelter.bsky.social · 18/08/2025
SARS-CoV-2 causes systemic vascular damage via viral proteins and immune dysregulation, driving acute organ failure and Long COVID. Vaccination reduces these risks significantly. davidlingenfelter.substack.com/p/the-vascul...
🧬 FOR IMMEDIATE RELEASE 

SARS-CoV-2 Confirmed as a Vascular Pathogen: New Report Details Mechanisms Behind Acute and Long-Term Damage

August 18, 2025 — El Cerrito, CA - A newly revised scientific report by David Lingenfelter, PhD, synthesizes global consensus and cutting-edge research to confirm that SARS-CoV-2 is not merely a respiratory virus—it is a systemic vascular pathogen. The report, originally generated via Google’s Gemini AI and refined through critique and reanalysis, offers a comprehensive account of the virus’s multifaceted assault on the vascular system.
🔍 Key Findings:
•	Endothelial Dysfunction as Central Pathology: SARS-CoV-2 disrupts vascular homeostasis via Spike protein–ACE2 interactions, Nucleocapsid protein–TLR2 signaling, and immune-mediated inflammation.
•	Microvascular Injury Drives Organ Failure: COVID-19-associated coagulopathy (CAC), microthrombosis, and vasculitis contribute to acute morbidity and mortality.
•	Long COVID Linked to Persistent Vascular Damage: Chronic endotheliopathy, fibrinaloid microclots, and senescent “zombie cells” underlie post-acute sequelae.
•	Vaccination Reduces Vascular Risk: Comparative data show significantly lower rates of cardiac complications post-vaccination versus post-infection.
🧠 Implications for Clinical Practice and Public Health:
•	COVID-19 should be classified as a systemic vascular disease.
•	Diagnostic protocols must incorporate endothelial biomarkers and advanced imaging.
•	Therapeutic strategies should target inflammation, thrombosis, and senescence.
•	Long-term surveillance is needed to assess cardiovascular risk in survivors.
📚 Future Directions: The report calls for randomized trials of senolytics, microclot-targeting therapies, and personalized vascular risk profiling. It also urges public health systems to prioritize vascular health as a pandemic preparedness measure.
🔗 Full report available on https://davidlingenfelter.substack.com/p/the-vascular-pathophysiology-of-sars-b2d
063
Reposted by Sonika Tyagi
Billy Hanlon @bhanlon15.bsky.social · 20/08/2025
SciTech Daily: 'Scientists Detect “Ghost” Proteins That Could Explain Long COVID' 'Scientists have uncovered viral protein fragments lingering in the blood of long COVID patients, potentially offering the first measurable biomarker for the condition.' scitechdaily.com/scientists-d...
scitechdaily.com
Scientists Detect “Ghost” Proteins That Could Explain Long COVID
Researchers discovered viral fragments in blood samples from long COVID patients, hinting at a possible biomarker. The findings suggest the virus may linger in hidden reservoirs, fueling ongoing sympt...
517566
Reposted by Sonika Tyagi
Healthcare-Global @healthcare-global.bsky.social · 27/08/2025
www.miragenews.com/gaps-in-mate...
miragenews.com
Gaps In Maternal Health In Australia, USA, And UK
Global study reveals alarming gaps in maternal health - Getty ImagesWomen across Australia, the United States (USA) and the United Kingdom (UK) are
031
Sonika Tyagi @tsonika.bsky.social · 27/08/2025
Contributions to review on graph learning:
011
Reposted by Sonika Tyagi
The Lancet Microbe @lancetmicrobe.bsky.social · 15/08/2025
New comment A layered strategy for tackling antimicrobial resistance: the Swiss cheese model for policy, prevention, and engagement www.thelancet.com/journals/lan... #IDSky #ClinMicro #AMR #OpenAccess #OA
The antimicrobial resistance Swiss cheese model
31911
Reposted by Sonika Tyagi
Leandros Boukas @leandrosboukas.bsky.social · 08/08/2025
This looks like a creative use of transformers. Looking forward to diving into this. arxiv.org/abs/2508.04111
arxiv.org
Negative binomial regression and inference using a pre-trained transformer
Negative binomial regression is essential for analyzing over-dispersed count data in in comparative studies, but parameter estimation becomes computationally challenging in large screens requiring mil...
182
Sonika Tyagi @tsonika.bsky.social · 07/08/2025
We had a fantastic day for team building and a send-off lunch and marking our mid-season break. It was a lovely opportunity to come together and celebrate, with some intense competition on the bowling lanes! 😆 We will miss having you around at @tyagilab.bsky.social @muraliaadhitya.bsky.social 🙌🏽
031
Reposted by Sonika Tyagi
Nextflow @nextflow.io · 05/08/2025
🚀 Excited to see this work on "Empowering bioinformatics communities with Nextflow and @nf-co.re" is now published, showcasing the incredible growth and impact of the community. link.springer.com/article/10.1...
link.springer.com
Empowering bioinformatics communities with Nextflow and nf-core - Genome Biology
Standardized analysis pipelines contribute to making data bioinformatics research compliant with the paradigm of Findability, Accessibility, Interoperability, and Reusability (FAIR), and facilitate co...
11410
Reposted by Sonika Tyagi
Nature Methods @natmethods.nature.com · 04/08/2025
An analysis shows that current deep learning models do not beat linear baselines in predicting gene perturbation effects, thus emphasizing the importance of further method development and evaluation. @const-ae.bsky.social @wkhuber.bsky.social @s-anders.bsky.social www.nature.com/articles/s41...
nature.com
Deep-learning-based gene perturbation effect prediction does not yet outperform simple linear baselines - Nature Methods
The analysis presented in this Brief Communication shows that, despite their complexity, current deep learning models do not outperform linear baselines in predicting gene perturbation effects, thus e...
06425