Xiaowei Jiang
@johnjxw.bsky.social
Computational and evolutionary biologist, studying virus, microbial and cancer evolution. Views my own.
Reposted by Xiaowei Jiang
Simulation-based Inference (SBI) brings the power of generative modeling to parameter inference in science and engineering. We have now written a practical guide on how to effectively use these methods across domains 🔭👁️🗨️🦞
Simulation-based inference (SBI) has transformed parameter inference across a wide range of domains. To help practitioners get started and make the most of these methods, we joined forces with researchers from many institutions and wrote a practical guide to SBI.
📄 Paper: arxiv.org/abs/2508.12939
📄 Paper: arxiv.org/abs/2508.12939
Simulation-Based Inference: A Practical Guide
A central challenge in many areas of science and engineering is to identify model parameters that are consistent with prior knowledge and empirical data. Bayesian inference offers a principled framewo...
arxiv.org
November 21, 2025 at 3:17 PM
Simulation-based Inference (SBI) brings the power of generative modeling to parameter inference in science and engineering. We have now written a practical guide on how to effectively use these methods across domains 🔭👁️🗨️🦞
It works as we last tried sequences from an unreleased glycoprotein-host receptor complex, and it predicted a positive interaction score!
Our PLM-interact is out in Nature Communications! We show that jointly encoding protein pairs using protein language models improves protein–protein interaction prediction performance and enables fine-tuning to predict mutation effects in human PPIs. www.nature.com/articles/s41...
PLM-interact: extending protein language models to predict protein-protein interactions - Nature Communications
Protein structure can be predicted from amino acid sequences with unprecedented accuracy, yet the prediction of protein–protein interactions remains a challenge. Here, authors present a sequence-based...
www.nature.com
November 21, 2025 at 12:38 AM
It works as we last tried sequences from an unreleased glycoprotein-host receptor complex, and it predicted a positive interaction score!
Transmissible vaccines, gene drives... and releasing these to ecosystems, oh man: doi.org/10.1073/pnas...
Gene drives, species complexes, and the risks of collateral damage | PNAS
Gene drives, species complexes, and the risks of collateral damage
doi.org
November 17, 2025 at 6:23 PM
Transmissible vaccines, gene drives... and releasing these to ecosystems, oh man: doi.org/10.1073/pnas...
Ok, I am gonna watch some cat clips to cheer me up...
Can't emphasize how damaging this is. A single non-scientist—Peter Bogner—holds all power & makes all decisions at GISAID & provides no justifications for any of them, except blatantly false ones.
He has that power because he conned rich & powerful people into giving it to him. Enough.
He has that power because he conned rich & powerful people into giving it to him. Enough.
I want to spell this out in case the implications aren't clear:
This means all public tools/webapps of GISAID data (all the ones you've been used to seeing thru the pandemic, as far as we can tell) are prohibited.
The file allowed this. Cut that - cut off all tools the public & others were using.
This means all public tools/webapps of GISAID data (all the ones you've been used to seeing thru the pandemic, as far as we can tell) are prohibited.
The file allowed this. Cut that - cut off all tools the public & others were using.
November 16, 2025 at 11:29 AM
Ok, I am gonna watch some cat clips to cheer me up...
Useful concepts, when everything is mixing with everything these days or 4.5 billion years ago...
I've often wondered about what we should call organisms whose similarity might be due to acquired genetic material. It got a little complicated, but I made a stab at it here
Classifying Convergences in the Light of Horizontal Gene Transfer: Epaktovars and Xenotypes academic.oup.com/mbe/article/...
Classifying Convergences in the Light of Horizontal Gene Transfer: Epaktovars and Xenotypes academic.oup.com/mbe/article/...
Classifying Convergences in the Light of Horizontal Gene Transfer: Epaktovars and Xenotypes
Abstract. The classification of living systems presents significant challenges due to the prevalence of gene transfer between genomes. Traditional taxonomi
academic.oup.com
November 15, 2025 at 1:45 PM
Useful concepts, when everything is mixing with everything these days or 4.5 billion years ago...
James Kirk! And lots of interesting research: royalsociety.org/news/2025/11...
November 14, 2025 at 3:21 PM
James Kirk! And lots of interesting research: royalsociety.org/news/2025/11...
Reposted by Xiaowei Jiang
Modestly, we decided to review a century's worth of theories of balancing selection :) onlinelibrary.wiley.com/doi/10.1111/...
A century of theories of balancing selection
Traits that affect organismal fitness are often highly genetically variable. This genetic variation is vital for populations to adapt to their environments, but it is also surprising given that natur...
onlinelibrary.wiley.com
November 14, 2025 at 2:31 PM
Modestly, we decided to review a century's worth of theories of balancing selection :) onlinelibrary.wiley.com/doi/10.1111/...
One thing I do not understand is that why people keep proposing new concepts for already confusing concepts?
press.uchicago.edu/ucp/books/bo...
press.uchicago.edu/ucp/books/bo...
Nonadaptive Selection
The first comprehensive explanation of a widely applicable but underappreciated mechanism of evolution operating at higher levels of organization than the individual. In this important treatise, eco...
press.uchicago.edu
November 12, 2025 at 12:25 PM
One thing I do not understand is that why people keep proposing new concepts for already confusing concepts?
press.uchicago.edu/ucp/books/bo...
press.uchicago.edu/ucp/books/bo...
This is quite interesting, they used a conservative method instead of allele frequency (not there yet maybe), but then it is good enough for the questions they asked. www.cell.com/cell/fulltex...
November 12, 2025 at 12:22 PM
This is quite interesting, they used a conservative method instead of allele frequency (not there yet maybe), but then it is good enough for the questions they asked. www.cell.com/cell/fulltex...
Reposted by Xiaowei Jiang
🚨New preprint🚨
In a multidisciplinary tour-de-force, @itingtu.bsky.social shows that, because #H5N1 flu is now routinely infecting seals and sealions in Peru, it's also spilling over into the vampire bats that feed from beaches...
🦭🦠🦇
(1/2)
In a multidisciplinary tour-de-force, @itingtu.bsky.social shows that, because #H5N1 flu is now routinely infecting seals and sealions in Peru, it's also spilling over into the vampire bats that feed from beaches...
🦭🦠🦇
(1/2)
My 1st first-author paper is out as a preprint! Excited to share the fascinating story of marine-feeding vampire bats 🌊🦇 We found that marine-diet bats were exposed to #H5#AvianFlu 🦠 from marine wildlife and could potentially spread it to livestock via blood feed🩸
doi.org/10.1101/2025.11.09.686930
doi.org/10.1101/2025.11.09.686930
November 12, 2025 at 11:33 AM
🚨New preprint🚨
In a multidisciplinary tour-de-force, @itingtu.bsky.social shows that, because #H5N1 flu is now routinely infecting seals and sealions in Peru, it's also spilling over into the vampire bats that feed from beaches...
🦭🦠🦇
(1/2)
In a multidisciplinary tour-de-force, @itingtu.bsky.social shows that, because #H5N1 flu is now routinely infecting seals and sealions in Peru, it's also spilling over into the vampire bats that feed from beaches...
🦭🦠🦇
(1/2)
Later SARS-CoV-2 viruses in humans evolved the ability to use New World monkey ACE2, and recombination seems to be able to sample more mutations for the spike gene to cross species barriers.
doi.org/10.1093/molb...
doi.org/10.1093/molb...
SARS-CoV-2 Evolution in Humans Enables its Transmission to Nonhuman Primates
Abstract. Zoonotic viruses frequently cross species barriers, but the molecular processes enabling reverse zoonosis remains poorly defined. The COVID-19 pa
doi.org
November 10, 2025 at 11:45 AM
Later SARS-CoV-2 viruses in humans evolved the ability to use New World monkey ACE2, and recombination seems to be able to sample more mutations for the spike gene to cross species barriers.
doi.org/10.1093/molb...
doi.org/10.1093/molb...
Reposted by Xiaowei Jiang
Wild story. Global spread of African Swine Flu is said to be depriving Amur tigers of their wild boar prey in remote Siberia. The swine flu link looks conjectural but plausible in this account.
Another case of panzootic consequences?
Another case of panzootic consequences?
‘The tigers are hungry’: endangered but deadly, the world’s largest big cat is sowing fear in Siberia’s villages
‘The tigers are hungry’: endangered but deadly, the world’s largest big cat is sowing fear in Siberia’s villages
The spread of African swine flu among the wild boars the animals eat has led to the deadliest winter for attacks on people in the Russian region for decades – and a spike in tiger killings
www.theguardian.com
November 10, 2025 at 10:36 AM
Wild story. Global spread of African Swine Flu is said to be depriving Amur tigers of their wild boar prey in remote Siberia. The swine flu link looks conjectural but plausible in this account.
Another case of panzootic consequences?
Another case of panzootic consequences?
Reposted by Xiaowei Jiang
Here is a formal impossibility result for XAI: Informative Post-Hoc Explanations Only Exist for Simple Functions. I'll give an online presentation about this work next tuesday in @timvanerven.nl 's Theory of Interpretable AI Seminar:
arxiv.org/abs/2508.11441
tverven.github.io/tiai-seminar/
arxiv.org/abs/2508.11441
tverven.github.io/tiai-seminar/
November 7, 2025 at 6:25 AM
Here is a formal impossibility result for XAI: Informative Post-Hoc Explanations Only Exist for Simple Functions. I'll give an online presentation about this work next tuesday in @timvanerven.nl 's Theory of Interpretable AI Seminar:
arxiv.org/abs/2508.11441
tverven.github.io/tiai-seminar/
arxiv.org/abs/2508.11441
tverven.github.io/tiai-seminar/
Reposted by Xiaowei Jiang
Delighted that our review on resistance to last-resort antibiotics in Enterococci has been published in FEMS Microbiology Reviews
academic.oup.com/femsre/advan... (accepted manuscript version)
TL;DR: the Enterococci are really good at evolving resistance to antibiotics in new and creative ways.
academic.oup.com/femsre/advan... (accepted manuscript version)
TL;DR: the Enterococci are really good at evolving resistance to antibiotics in new and creative ways.
Resistance to last-resort antibiotics in enterococci
Abstract. The genus Enterococcus comprises a diverse group of species, many of which are commensal members of the gut microbiota of humans and animals. The
academic.oup.com
November 8, 2025 at 5:41 PM
Delighted that our review on resistance to last-resort antibiotics in Enterococci has been published in FEMS Microbiology Reviews
academic.oup.com/femsre/advan... (accepted manuscript version)
TL;DR: the Enterococci are really good at evolving resistance to antibiotics in new and creative ways.
academic.oup.com/femsre/advan... (accepted manuscript version)
TL;DR: the Enterococci are really good at evolving resistance to antibiotics in new and creative ways.
There are some good comments relevant to how computational biologists use coder LLMs.
blog.genesmindsmachines.com/p/llms-excel...
blog.genesmindsmachines.com/p/llms-excel...
LLMs excel at programming—how can they be so bad at it?
My explanation for the mystery of why LLMs can be both exceptionally good and quite terrible at programming.
blog.genesmindsmachines.com
November 9, 2025 at 12:17 PM
There are some good comments relevant to how computational biologists use coder LLMs.
blog.genesmindsmachines.com/p/llms-excel...
blog.genesmindsmachines.com/p/llms-excel...
This one is interesting, basically consistent with my experiments.
www.science.org/content/blog...
www.science.org/content/blog...
Can ChatGPT help science writers?
www.science.org
November 9, 2025 at 12:13 PM
This one is interesting, basically consistent with my experiments.
www.science.org/content/blog...
www.science.org/content/blog...
"340 years and they still haven't figured out their audio setup." LOL
www.youtube.com/watch?v=db36...
www.youtube.com/watch?v=db36...
The Honorary Patronage of Stephen Colbert II University Philosophical Society
YouTube video by TCDPhil
www.youtube.com
November 6, 2025 at 2:37 PM
"340 years and they still haven't figured out their audio setup." LOL
www.youtube.com/watch?v=db36...
www.youtube.com/watch?v=db36...
Sequenced genomes from a recent EVD outbreak "form a well-supported phylogenetic cluster with genomes from the 1976 Yambuku/Mayinga outbreak"!
www.researchsquare.com/article/rs-7...
www.researchsquare.com/article/rs-7...
Phylogenetic analysis of initial Ebola virus genomes from the 16th Ebola Virus Disease Outbreak in the Democratic Republic of Congo, September 2025
On 4 September 2025, the Ministry of Public Health, Hygiene and Social Welfare officially declared the 16th Ebola virus disease (EVD) outbreak in the Democratic Republic of the Congo (DRC). As of 13 O...
www.researchsquare.com
November 3, 2025 at 3:39 PM
Sequenced genomes from a recent EVD outbreak "form a well-supported phylogenetic cluster with genomes from the 1976 Yambuku/Mayinga outbreak"!
www.researchsquare.com/article/rs-7...
www.researchsquare.com/article/rs-7...
Lucky, I was not drinking my morning coffee while reading this.
November 2, 2025 at 12:43 PM
Lucky, I was not drinking my morning coffee while reading this.
Reposted by Xiaowei Jiang
As expected, unfortunately.
If ever you needed a reason for never using GISAID ever again (as a data producer or data user - we're both), look no further.
Time to move on to more trusted and transparent solutions.
If ever you needed a reason for never using GISAID ever again (as a data producer or data user - we're both), look no further.
Time to move on to more trusted and transparent solutions.
October 31, 2025 at 1:41 AM
As expected, unfortunately.
If ever you needed a reason for never using GISAID ever again (as a data producer or data user - we're both), look no further.
Time to move on to more trusted and transparent solutions.
If ever you needed a reason for never using GISAID ever again (as a data producer or data user - we're both), look no further.
Time to move on to more trusted and transparent solutions.
"...spike protein of this novel bat coronavirus possesses a functional furin cleavage site at the S1/S2 junction with a unique amino acid sequence motif (RDAR) that differs from that found in SARS-CoV-2 (RRAR) by only one amino acid..."
www.biorxiv.org/content/10.1...
www.biorxiv.org/content/10.1...
A divergent betacoronavirus with a functional furin cleavage site in South American bats
Bats are natural reservoirs for a wide range of RNA viruses. Members of the genus Betacoronavirus, including Severe Acute Respiratory Syndrome virus 2 (SARS-CoV-2) and Middle East Respiratory Syndrome...
www.biorxiv.org
October 28, 2025 at 5:48 PM
"...spike protein of this novel bat coronavirus possesses a functional furin cleavage site at the S1/S2 junction with a unique amino acid sequence motif (RDAR) that differs from that found in SARS-CoV-2 (RRAR) by only one amino acid..."
www.biorxiv.org/content/10.1...
www.biorxiv.org/content/10.1...
Interesting paper: "...no clear association between genomic or phenotypic traits and recombination rate...", but "the devil is in the details"
doi.org/10.1073/pnas...
doi.org/10.1073/pnas...
Evolution of homologous recombination rates across bacteria | PNAS
Bacteria are nonsexual organisms but are capable of exchanging DNA at diverse degrees
through homologous recombination. Intriguingly, the rates of ...
doi.org
October 26, 2025 at 11:26 AM
Interesting paper: "...no clear association between genomic or phenotypic traits and recombination rate...", but "the devil is in the details"
doi.org/10.1073/pnas...
doi.org/10.1073/pnas...
When Apple iCloud relay is turned on, you use Safari to browse Google scholar, Google blocks you. But when you switch to Google Chrome, no block... hmm, evil? hmm...
October 26, 2025 at 11:11 AM
When Apple iCloud relay is turned on, you use Safari to browse Google scholar, Google blocks you. But when you switch to Google Chrome, no block... hmm, evil? hmm...
Reposted by Xiaowei Jiang
Turns out NLP is just vision
Z.ai released a paper very similar to DeepSeek-OCR on the same exact day (a few hours earlier afaict)
Glyph is just a framework, not a model, but they got Qwen3-8B (128k context) to handle over 1 million context by rendering input as images
arxiv.org/abs/2510.17800
Glyph is just a framework, not a model, but they got Qwen3-8B (128k context) to handle over 1 million context by rendering input as images
arxiv.org/abs/2510.17800
October 21, 2025 at 4:39 PM
Turns out NLP is just vision
It is scandalously expensive now. Not to mention if you include your family. Let's hope British science can back it up in the long run.
Gulf widens between UK and rival nations on scientist visa costs
Analysis for Royal Society finds UK is now 22 times pricier upfront than similar countries
www.researchprofessionalnews.com/rr-news-uk-c...
Analysis for Royal Society finds UK is now 22 times pricier upfront than similar countries
www.researchprofessionalnews.com/rr-news-uk-c...
Cost gap widens between UK and other countries for scientist visas - Research Professional News
Analysis for Royal Society finds UK is now 22 times pricier upfront than similar countries
www.researchprofessionalnews.com
October 21, 2025 at 2:27 PM
It is scandalously expensive now. Not to mention if you include your family. Let's hope British science can back it up in the long run.