Sign in

Alexander Rakitko

@alexrakitko.bsky.social
18 followers 26 following 38 posts

Genetic testing, population genetics, multifactor diseases, ML in HealthCare, Ph.D.

PostsRepliesMedia
Alexander Rakitko @alexrakitko.bsky.social · 13/05/2026
Triple-negative breast cancer is not one disease, but a set of distinct cellular ecosystems. And the structure of that ecosystem may help predict whether chemotherapy will work. www.nature.com/articles/s41... #Cancer #SpatialTranscriptomics #RNAseq
000
Alexander Rakitko @alexrakitko.bsky.social · 13/05/2026
Tumors that responded better to chemotherapy more often showed: - Interferon signaling in cancer cells — a sign of immune alert. - Higher HLA class II expression — potentially making tumor cells more visible to the immune system. - More actively dividing cells, especially in S phase
100
Alexander Rakitko @alexrakitko.bsky.social · 13/05/2026
Triple-negative breast cancer is an aggressive breast cancer subtype lacking estrogen, progesterone, and HER2 receptors, which limits targeted treatment options and keeps chemotherapy central.
100
Alexander Rakitko @alexrakitko.bsky.social · 13/05/2026
🧬Can chemotherapy response in aggressive breast cancer be predicted from the tumor “ecosystem”? New @nature.com study shows that single-cell RNAseq and spatial transcriptomics can predict response to chemotherapy in triple-negative breast cancer (model reaching AUC = 0.84).
100
Alexander Rakitko @alexrakitko.bsky.social · 13/05/2026
Their ancestry is most closely related to populations from Central America and northern South America, although no sampled mainland group is an exact match. These early Caribbean communities were very small and locally structured, with little evidence of close-kin mating.
000
Alexander Rakitko @alexrakitko.bsky.social · 13/05/2026
Using genome-wide data from 19 individuals from Hispaniola’s Samaná Peninsula, the authors show that pre-Ceramic populations across Hispaniola and Cuba likely came from a single ancestry source and maintained genetic continuity for thousands of years.
100
Alexander Rakitko @alexrakitko.bsky.social · 13/05/2026
🧬New preprint from David Reich and colleagues examines the population history of the first Caribbean inhabitants, which has been difficult to reconstruct because very few early human remains are known. www.biorxiv.org/content/10.6... #PopulationGenetics #CaribbeanIslands #aDNA
100
Alexander Rakitko @alexrakitko.bsky.social · 11/05/2026
This is not the first study to suggest this idea, but it goes much further than earlier work and gives it much stronger support. www.cell.com/cell/fulltex... #Polyploidy #Evolution #Plants #MolecularClocks #Phylogenie
000
Alexander Rakitko @alexrakitko.bsky.social · 11/05/2026
But during catastrophes such as global cooling, warming, and mass extinctions, this genetic excess can become useful. Extra gene copies give plants more raw material for adaptation.
100
Alexander Rakitko @alexrakitko.bsky.social · 11/05/2026
Polyploidy is not rare in plants. Bananas often have three sets of chromosomes, while wheat can have up to six. In stable times, extra chromosome sets can be a burden. The genome is larger, cells are more expensive to maintain, and fertility may decline.
100
Alexander Rakitko @alexrakitko.bsky.social · 11/05/2026
One example is the Cretaceous-Paleogene boundary around 66 million years ago, when an asteroid impact was linked to the extinction of non-avian dinosaurs and a major crisis across ecosystems.
100
Alexander Rakitko @alexrakitko.bsky.social · 11/05/2026
Using molecular-clock dating and 44 plant fossil calibrations, the authors show that ancient genome duplications often cluster around periods of environmental crisis.
100
Alexander Rakitko @alexrakitko.bsky.social · 11/05/2026
🧬When the world is shaken by disasters on a planetary scale, it may be time to duplicate the genome💥🌍🌱 In a new paper in @cellcellpress.bsky.social, researchers analyzed 470 genomes of flowering plants and suggest that whole-genome duplication may act as an evolutionary emergency mode. 🧵⤵️
120
Alexander Rakitko @alexrakitko.bsky.social · 09/05/2026
This matters because such findings affect how we understand early microbiome development. Another example is the TwinsUK cohort, where CroCoDeEL identified 202 contaminated samples. Many shared the same sources, had higher species richness, and were more similar to each other than expected.
020
Alexander Rakitko @alexrakitko.bsky.social · 09/05/2026
The authors applied the tool to published datasets and found previously unnoticed contamination. In infant microbiome studies, some bacteria interpreted as possible mother-to-child transmission were likely contamination from maternal samples.
120
Alexander Rakitko @alexrakitko.bsky.social · 09/05/2026
In many ways, modern bioinformatics is the art of quality control. Samples from relatives — for example, a mother and her child — may naturally end up in neighboring wells on a plate. In that case, cross-contamination can be mistaken for microbiome heritability.
120
Alexander Rakitko @alexrakitko.bsky.social · 09/05/2026
If a drop from a Bloody Mary splashes into a Moscow Mule, the second glass gets a small, diluted but still recognizable trace of the first cocktail. CroCoDeEL looks for these suspicious “drops” in metagenomic profiles.
110
Alexander Rakitko @alexrakitko.bsky.social · 09/05/2026
🧬CroCoDeEL is a new tool for detecting cross-sample contamination in metagenomic data.🐊🦠 The idea behind CroCoDeEL is simple and elegant. Imagine different cocktails poured into a row of glasses...🧵 www.nature.com/articles/s41... #Microbiome #Metagenomics #Contamination
2177
Alexander Rakitko @alexrakitko.bsky.social · 09/05/2026
Seems so! I’d love to see more microbiome data annotated with geographic and temporal metadata. It would be interesting to know whether we can infer something about an individual’s life history or places of residence from their microbiome.
000
Alexander Rakitko @alexrakitko.bsky.social · 09/05/2026
Using 6,783 metagenomes, the authors also found 178 commensal sweep clusters associated with colorectal cancer, IBD and type 2 diabetes. Practically, these signals can be invisible at the species level. Different sweep clusters within the same species-level group may map to different host diseases.
000
Alexander Rakitko @alexrakitko.bsky.social · 09/05/2026
This matters because microbiome studies often work at the species level, but the biologically relevant unit may be smaller: an ecological population created by a sweep. Interesting that many commensal sweep clusters appeared on multiple continents and some may have spread globally within decades.
100
Alexander Rakitko @alexrakitko.bsky.social · 09/05/2026
The authors show that genome-wide selective sweeps are widespread across gut bacteria. When one clone gets a useful adaptation, its whole genome can hitchhike with it, outcompeting related lineages. The clone then diversifies again, leaving a “broom-like” phylogenetic pattern.
100
Alexander Rakitko @alexrakitko.bsky.social · 09/05/2026
🧬New @Nature paper suggests that the human gut microbiome may be shaped by evolutionary events that look surprisingly similar to global epidemics.🦠🧵 www.nature.com/articles/s41... #Microbiome #Genomics #Evolution #PopulationGenetics
421
Alexander Rakitko @alexrakitko.bsky.social · 09/05/2026
Genes involved in toxin metabolism — On land, toxins help plants defend themselves. In water, the threats are different, so some of these genes became less necessary.
000
Alexander Rakitko @alexrakitko.bsky.social · 09/05/2026
Immune defense genes — Some land pathogens don’t survive in water, so water lilies likely needed fewer resistance genes.
100
Alexander Rakitko @alexrakitko.bsky.social · 09/05/2026
Genes involved in cell wall development — On land, plants need strong tissues to stand upright. In water, buoyancy helps support them, so some of these genes became less necessary.
100
Alexander Rakitko @alexrakitko.bsky.social · 09/05/2026
🧬Cool paper on the evolution of water lilies.🌺 What I liked most was how they adapted to aquatic life. The authors found 481 genes that water lilies have lost, including: rdcu.be/fhdEb #WaterLily #Evolution #Genetics #Adaptation
rdcu.be
Water lily complete genomes illuminate the innovations of water lilies and early angiosperms
Nature Plants - This study decodes gapless genomes for three water lily species, discovering genetic keys to fertilization, floral colour and scent, and offering a molecular roadmap for how early...
110
Alexander Rakitko @alexrakitko.bsky.social · 01/05/2026
The optimum appears to be a target with a strong enough biological signal, but without an excessively broad phenotypic footprint.
000
Alexander Rakitko @alexrakitko.bsky.social · 01/05/2026
Perhaps the most important result is that the association between pleiotropy and drug-target success is non-linear. Too little pleiotropy may indicate weak or narrow biological involvement. Too much pleiotropy may imply systemic effects and safety liabilities.
110
Alexander Rakitko @alexrakitko.bsky.social · 01/05/2026
This illustrates why such studies are especially important: even before clinical trials, we can start to account for potential safety liabilities. Overall GWAS support was associated with a higher chance of clinical success — roughly a threefold increase.
100
Alexander Rakitko @alexrakitko.bsky.social · 01/05/2026
But there were also opposite examples. For instance, the APOE p.Cys130Arg variant was associated with lower risk of age-related macular degeneration and non-alcoholic fatty liver disease, but higher risk of Alzheimer’s disease.
100
Alexander Rakitko @alexrakitko.bsky.social · 01/05/2026
As expected, in most cases the effects of pleiotropic lead variants were concordant: 92.5% showed the same direction of association across diseases. In other words, the genetic variant tended to increase the risk of the associated diseases.
100
Alexander Rakitko @alexrakitko.bsky.social · 01/05/2026
This is highly important for drug discovery, because a therapeutic target is usually embedded in a network of molecular and biological mechanisms rather than linked to a single isolated outcome.
100
Alexander Rakitko @alexrakitko.bsky.social · 01/05/2026
The authors show that most genes are pleiotropic — meaning that the same gene or genetic variant can be associated with multiple traits or diseases. Among 8,285 disease-associated genes, about 64% were linked to more than one disease.
100
Alexander Rakitko @alexrakitko.bsky.social · 01/05/2026
🧬New massive study of more than one trillion variant–trait associations suggests that drug-target success may lie in the middle ground between low and high pleiotropy. 💊 🧵 platform.opentargets.org www.biorxiv.org/content/10.6... #GWAS #OpenTargets #Pleiotropy #DrugDesign #ComplexDiseases
100
Alexander Rakitko @alexrakitko.bsky.social · 30/04/2026
Even more concerning: when users included a wrong assumption, polite models were less likely to correct it. Could a too-polite AI hide serious diagnoses - like cancer - just to avoid upsetting you?
000
Alexander Rakitko @alexrakitko.bsky.social · 30/04/2026
Bitter truth or sweet AI lie? A new @Nature paper shows that making LLMs more “warm” and polite can actually make them less accurate. Warm-tuned models made +8.6% more errors on medical questions. And the effect got much worse when the user was emotionally vulnerable. www.nature.com/articles/s41...
nature.com
Training language models to be warm can reduce accuracy and increase sycophancy - Nature
Experiments on five different language models show that training language models to produce warmer responses can undermine the accuracy of their output, especially when users express feelings of sadne...
100
Reposted by Alexander Rakitko
Shai Carmi @shaicarmi.bsky.social · 28/04/2026
A new preprint! A short case report in forensic genetics. Skeletal remains discovered in a cave in central Israel were thought to match a person of Ashkenazi Jewish origin missing since 1948. The authorities were confident about the identification, but... 1/3 www.biorxiv.org/content/10.6...
biorxiv.org
192
Alexander Rakitko @alexrakitko.bsky.social · 30/04/2026
🧬A new @Nature study of 258 ancient genomes from Germany shows that Early Medieval southern Germany was not formed by a simple barbarian replacement of the Roman population.🏺 www.nature.com/articles/s41... #aDNA #RomanEmpire #Admixture #PCA #PopulationGenetics #PedigreeReconstruction
nature.com
Demography and life histories across the Roman frontier in Germany 400–700 ce - Nature
Analysis of 258 ancient genomes from southern Germany reveals a major demographic shift during the late fifth century, yet family structures persisted from Late Roman times, demonstrating cultural con...
0155
Reposted by Alexander Rakitko
Hans Fredrik Sunde @hfsunde.bsky.social · 06/06/2025
Our paper on indirect assortative mating is now out in @natcomms.nature.com! In it, we provide refined definitions of terms used to explain partner similarity, develop statistical models, and find evidence of surprisingly high social homogamy for education. Link: doi.org/10.1038/s414...
1328
Reposted by Alexander Rakitko
Mike Inouye @mikeinouye.bsky.social · 05/06/2025
📣 Latest from the lab: Performance of deep-learning-based approaches to improve polygenic scores www.nature.com/articles/s41... Its thought deep learning will substantially improve PGS but the reality is MANY have tried but no/little gain has been seen so far. Here we report our negative results.
419682
Reposted by Alexander Rakitko
Mike Inouye @mikeinouye.bsky.social · 22/03/2025
Sparse haplotype-based fine-scale local ancestry inference at scale reveals recent selection on immune responses www.nature.com/articles/s41...
nature.com
Sparse haplotype-based fine-scale local ancestry inference at scale reveals recent selection on immune responses - Nature Communications
Here, the authors present efficient local ancestry inference algorithms to analyse the population of origin for each individual and variant in the UK Biobank, improving controls for population structu...
0209
Reposted by Alexander Rakitko
Stephen Turner @stephenturner.us · 04/03/2025
Investigating the performance of Oxford Nanopore long-read sequencing with respect to Illumina microarrays and short-read sequencing www.biorxiv.org/content/10.1... 🧬🖥️🧪 Nextflow github.com/renatosantos...
0164
Reposted by Alexander Rakitko
Shai Carmi @shaicarmi.bsky.social · 18/02/2025
Happy to share a new preprint, to which I provided a minor contribution. Using genetic data from ~100k human embryos, we found ~1k that were haploid or triploid. We describe a thorough characterization of these embryos, with many interesting findings 🧵 1/9 www.biorxiv.org/content/10.1...
1177
Reposted by Alexander Rakitko
Andrea Ganna @andganna.bsky.social · 07/02/2025
The long awaited European guidelines for the use of Polygenic score for primary prevention of cardiovascular diseases are out! It took > 10 years from the first results proposing its use as additional screening factor among patients with intermediate CVD risk academic.oup.com/eurheartj/ad...
academic.oup.com
Clinical utility and implementation of polygenic risk scores for predicting cardiovascular disease: A clinical consensus statement of the ESC Council on Cardiovascular Genomics, the ESC Cardiovascular...
Abstract. Genome-wide association studies have revealed hundreds of genetic variants associated with cardiovascular diseases (CVD). Polygenic risk scores (
02810
Reposted by Alexander Rakitko
Andrea Ganna @andganna.bsky.social · 08/01/2025
Did you know you can get DNA viruses from standard blood WGS done in large biobanks? (i didn’t). One point in favor of WGS vs WES Yet another nice paper from Yuki’s group www.nature.com/articles/s41...
nature.com
Blood DNA virome associates with autoimmune diseases and COVID-19 - Nature Genetics
Analysis of the blood DNA virome in patients with COVID-19 and autoimmune disease associates endogenous HHV-6 (eHHV-6) and high anellovirus load with increased disease risk, most notably for systemic ...
1144
Reposted by Alexander Rakitko
Jonas Meisner @rosemeis.bsky.social · 07/01/2025
I am excited that our *hapla* framework for fast and scalable haplotype clustering in phased genotype data has been published in Nature Communications! GitHub: github.com/Rosemeis/hapla Paper: www.nature.com/articles/s41...
nature.com
Leveraging haplotype information in heritability estimation and polygenic prediction - Nature Communications
Here the authors develop a framework to leverage haplotype information through clustering for heritability estimation and polygenic prediction. Their approach estimates disease risk more accurately th...
154
Reposted by Alexander Rakitko
Shai Carmi @shaicarmi.bsky.social · 30/12/2024
Nice analysis by PRS company Genomics plc, quantifying the potential reduction in mortality due to PRS screening. Assuming people at high risk for 7 diseases get early access to prevention/detection, they estimate saving 95 lives for each 100k people. www.medrxiv.org/content/10.1...
1104
Reposted by Alexander Rakitko
Marios Georgakis @mariosgeorgakis.bsky.social · 22/12/2024
New metabolomic aging clock trained on 168 metabolites in UK Biobank (N=225,212)⏲️ 👉R2 up to 0.35 👉associations with frailty index, telomere length, self-rated health 👉prediction of all-cause mortality 1/3
195
Reposted by Alexander Rakitko
Marios Georgakis @mariosgeorgakis.bsky.social · 17/12/2024
It's the era of large-scale biobanks❗️ 🧬GWAS summary stats from meta-analyses of: 👉MVP (N=630K) 👉FinnGen (N=420K) 👉UK Biobank (N=500K) ...for 330 clinical outcomes Unparalleled resource for phenome-wide explorations of genetic variation🔗: mvp-ukbb.finngen.fi
0101