Sandra Álvarez
LightNews LightNews
banner
sabifo4.bsky.social
Sandra Álvarez
@sabifo4.bsky.social
770 followers 480 following 43 posts
Researcher at UCL (CLOE) | Evol Biol & Comp Methods | Web Manager, Treasurer, and Blog Co-EIC at SRUK/CERU | Ashtangi & pianist | She/Her
Website: https://sabifo4.github.io/
GitHub: https://github.com/sabifo4
Linktree: https://linktr.ee/sabifo4
Posts Replies Media Videos
sabifo4.bsky.social
Back from this great workshop at Beijing Normal University! Thanks to everyone who worked so hard to bring this event to life, it has been a great success!💫 Special thanks to Cai-jin Chen for climbing the Great Wall with me, truly unforgettable! 谢谢🌹
Cai-jin and I before climbing the Great Wall Cai-jin and I at the highest peak of the Great Wall
November 12, 2025 at 1:32 PM
sabifo4.bsky.social
I am excited to announce that I am back to UCL (CLOE) to work with Ziheng Yang, Mario dos Reis, and Phil Donoghue on Bayesian techniques for inferring timetrees with large genomic datasets and ancient genomes! 💻🧬🤩
Renovated entrance to the 4th floor in the Darwin building with the new CLOE logo
August 12, 2024 at 10:44 AM
sabifo4.bsky.social
Bacterial evolution, GOE, clock dating, machine learning... What is not to like!! Go and talk to Lénárd today at 17.45h (poster L9) and join him tomorrow at 11.45h (room 520E) to learn more about how compositionally constrained sites can drive LBA! #Evol2024
Lénárd Szhanto's poster (L9): Geological timescale of bacterial evolution and oxygen adaptation. Come meet him today at 17.45h (Evolution meeting)!
July 29, 2024 at 7:34 PM
sabifo4.bsky.social
Dating analyses based on single genes, concatenations that excluded each gene in turn, or analyses under a complex substitution model returned compatible timescales to those we had initially reported. E.g.: estimates for LUCA nodes (Ext Dat 2) (16/n)
Extended Data Fig. 2 Comparison of the posterior time estimates and confidence intervals for the two duplicates of LUCA inferred under the main calibration strategy cross-bracing A with the concatenated dataset and with the datasets for the three additional sensitivity analyses. Dots refer to estimated mean divergence times and triangles to estimated 2.5% and 97.5% quantiles. Straight lines are used to link the mean divergence times estimated in the same analysis under the two different relaxed-clock models (GBM and ILN). Labels in the x axis are informative about the clock model under which the analysis ran and the type of analysis we carried (see abbreviations below). Coloured dots are used to identify which time estimates were inferred when using the same dataset and strategy under GBM and ILN, while triangles refer to the corresponding upper and lower quantiles for the 95% confidence interval.
July 12, 2024 at 12:49 PM
sabifo4.bsky.social
You can also find the main output files generated when sampling from the prior and the posterior under directories "sum_files_[prior|posterior]" in each "*MCMCtree" directory in the GitHub repository (pic shows an example of one of those). (11/n)
github.com/sabifo4/LUCA...
Example showing where directories "sum_files_post" and "sum_files_prior" can be found in the GitHub repository under one of the "*MCMCtree*" directories, "02_MCMCtree".
July 12, 2024 at 12:46 PM
sabifo4.bsky.social
Dating analyses based on != partitioning schemes & clock models returned compatible timescales (Fig S7-9). When cross-bracing was not enabled, posterior time estimates for mirror nodes were much more uncertain (Fig S9) -- using cross-bracing in such scenarios is important! (9/n)
Supplemental Figure 7. Timetree inferred under a Bayesian node-dating approach with cross-bracing A. This figure shows the corresponding posterior time densities of the mirrored nodes for the last universal, archaeal, bacterial, and eukaryotic common ancestors (LUCA, LACA, LBCA and LECA, respectively); as well as the last common ancestor of the mitochondrial lineage (Mito-LECA) and the last plastid-bearing common ancestor (LPCA). Each posterior density corresponds to the posterior time density of each of the analyses we ran under the calibration strategy “cross-bracing A”: (blue) partitioned dataset analysed under GBM, (purple) partitioned dataset analysed under ILN, (dark green) concatenated dataset analysed under GBM, (brown) concatenated dataset analysed under ILN. Note that all nodes that correspond to the same divergences were cross-braced during these analyses. Supplemental Figure 8. Timetree inferred under a Bayesian node-dating approach with cross-bracing B. This figure shows the corresponding posterior time densities of the mirrored nodes for the last universal, archaeal, bacterial, and eukaryotic common ancestors (LUCA, LACA, LBCA and LECA, respectively); as well as the last common ancestor of the mitochondrial lineage (Mito-LECA) and the last plastid-bearing common ancestor (LPCA). Each posterior density corresponds to the posterior time density of each of the analyses we ran under the calibration strategy “cross-bracing B”: (blue) partitioned dataset analysed under GBM, (purple) partitioned dataset analysed under ILN, (dark green) concatenated dataset analysed under GBM, (brown) concatenated dataset analysed under ILN. Note that only nodes that correspond to the same divergences for which there are fossil constraints were cross-braced during these analyses. Supplemental Figure 9. Timetree inferred under a Bayesian node-dating approach without cross-bracing. This figure shows the corresponding posterior time densities of the mirrored nodes for the last universal, archaeal, bacterial, and eukaryotic common ancestors (LUCA, LACA, LBCA and LECA, respectively); as well as the last common ancestor of the mitochondrial lineage (Mito-LECA) and the last plastid-bearing common ancestor (LPCA). Each posterior density corresponds to the posterior time density of each of the analyses we ran when cross-bracing was not enabled: (blue) partitioned dataset analysed under GBM, (purple) partitioned dataset analysed under ILN, (dark green) concatenated dataset analysed under GBM, (brown) concatenated dataset analysed under ILN. Note that we applied the same age constraint to those nodes that corresponded to the same speciation event. Given that such nodes are no longer cross-braced, the posterior time densities therefore differ.
July 12, 2024 at 12:45 PM
sabifo4.bsky.social
In addition, we ran other analyses to benchmark the effect that partitioning, cross-bracing, and relaxed-clock models could have on species divergence times estimation. The attached figure shows which directories correspond to which analysis (8/n)
github.com/sabifo4/LUCA...
Inside LUCA-divtimes/01_PAML… • Directories 00_CODEML + 02_MCMCtree: GBM/ILN + concatenated alignment + cross-bracing A (i.e., all nodes that correspond to the same divergences are cross-braced) | prior and posterior • Directories 00_CODEML + 03_MCMCtree_nonbraced: GBM/ILN + concatenated alignment + cross-bracing B (i.e., only nodes that correspond to the same divergences for which there are fossil constraints are cross-braced) | prior and posterior • Directories 00_CODEML + 04_MCMCtree_fossbraced: GBM/ILN + concatenated alignment + without cross-bracing (i.e., cross-bracing is not enabled) | prior and posterior • Directories 01_CODEML + 05_MCMCtree_part: GBM/ILN + partitioned alignment + cross-bracing A | prior and posterior • Directories 01_CODEML + 06_MCMCtree_nonbraced: GBM/ILN + partitioned alignment + cross-bracing B | prior and posterior • Directories 01_CODEML + 07_MCMCtree_fossbraced: GBM/ILN + partitioned alignment + without cross-bracing | prior and posterior
July 12, 2024 at 12:44 PM
sabifo4.bsky.social
The duplicated LUCA node age estimates we obtained under GBM and ILN with our partitioned dataset (GBM, 4.18–4.33 Ga; ILN, 4.09–4.32 Ga; Fig 1) fall within our composite age estimate for LUCA ranging from 3.94 Ga to 4.52 Ga, comparable to previous studies. (7/n)
Our results suggest that LUCA lived around 4.2 Ga, with a 95% confidence interval spanning 4.09–4.33 Ga under the ILN relaxed-clock model (orange) and 4.18–4.33 Ga under the GBM relaxed-clock model (teal). Under a cross-bracing approach, nodes corresponding to the same species divergences (that is, mirrored nodes) have the same posterior time densities. This figure shows the corresponding posterior time densities of the mirrored nodes for the last universal, archaeal, bacterial and eukaryotic common ancestors (LUCA, LACA, LBCA and LECA, respectively); the last common ancestor of the mitochondrial lineage (Mito-LECA); and the last plastid-bearing common ancestor (LPCA). Purple stars indicate nodes calibrated with fossils. Arc, Archaea; Bac, Bacteria; Euk, Eukarya.
July 12, 2024 at 12:43 PM
sabifo4.bsky.social
In our main analysis, we cross-braced all nodes that referred to the same divergences ("cross-bracing A") and analysed the partitioned dataset. We show the time estimates under the autocorrelated-rates (GBM) and independent-rates (ILN) models (Fig 1). (6/n)
Our results suggest that LUCA lived around 4.2 Ga, with a 95% confidence interval spanning 4.09–4.33 Ga under the ILN relaxed-clock model (orange) and 4.18–4.33 Ga under the GBM relaxed-clock model (teal). Under a cross-bracing approach, nodes corresponding to the same species divergences (that is, mirrored nodes) have the same posterior time densities. This figure shows the corresponding posterior time densities of the mirrored nodes for the last universal, archaeal, bacterial and eukaryotic common ancestors (LUCA, LACA, LBCA and LECA, respectively); the last common ancestor of the mitochondrial lineage (Mito-LECA); and the last plastid-bearing common ancestor (LPCA). Purple stars indicate nodes calibrated with fossils. Arc, Archaea; Bac, Bacteria; Euk, Eukarya.
July 12, 2024 at 12:42 PM
sabifo4.bsky.social
We used a new cross-bracing approach implemented in MCMCtree to cross-brace those nodes corresponding to the same species divergences. E.g., nodes a & b (also u & v) have the same label; equality constraints are enforced during the MCMC so that ta = tb, tu = tv (Fig S4). (4/n)
Supplemental Figure 4. A tree with labelled nodes to apply equality constraints (i.e., nodes that are cross-braced). By labelling nodes a and b using the same label (i.e., ‘#1’), and nodes u and v using the same label (i.e., ‘#2’), equality constraints are enforced when running the PAML program MCMCtree on the following node ages: ta = tb, tu = tv. Black dots indicate nodes which age is constrained by fossil calibrations.
July 12, 2024 at 12:41 PM