Frequently Asked Questions

Q: How do I cite the TimeTree database?

A: Kumar S, Suleski M, Craig JM, Kasprowicz AE, Sanderford M, Li M, Stecher G, Hedges SB (2022) TimeTree 5: An Expanded Resource for Species Divergence Times. Mol Biol Evol doi.org/10.1093/molbev/msac174

Earlier Versions
Version 4: Kumar S, Stecher G, Suleski M, Hedges SB (2017) TimeTree: A Resource for Timelines, Timetrees, and Divergence Times. Mol Biol Evol doi:10.1093/molbev/msx116

Version 3: Hedges SB, Marin J, Suleski M, Paymer M & Kumar S (2015) Tree of Life Reveals Clock-Like Speciation and Diversification. Mol Biol Evol 32: 835-845.

Version 2: Kumar S & Hedges SB (2011) TimeTree2: species divergence times on the iPhone. Bioinformatics 27:2023-2024.

Version 1: Hedges SB, Dudley J, & Kumar S (2006). TimeTree: A public knowledge-base of divergence times among organisms. Bioinformatics 22: 2971-2972.

Q: How did are TimeTree's molecular time estimates derived?

A: The molecular time estimates in Timetree represent a synthesis of published time estimates that were obtained from scientific literature. We present a detailed description of how we derive the time estimates in Kumar et al. (2022) but see also Hedges et al. (2015)

Q: How are TimeTree's summary times calculated?

A: We calculate summary time estimates using a simple average across all relevant time estimates for a given divergence. Additionally, we provide median times as some published estimates diverge substantially from others.

Q: What are adjusted times?

A: Due to conflicting time estimates between studies, ancestral nodes can be assigned younger ages than their descendants. A smoothing technique is used to adjust these times so the resulting tree is ultrametric, as described in Kumar et al. (2022).

Q: I found a species on a previous version of TimeTree but can't find it in TimeTree 5.

A: Previously, when data was unavailable to resolve a parent group, we would select a single member species as its representative tree tip. As of TimeTree 5, these groups now serve as their own representative tips. If you are trying to locate a missing species in TimeTree, we recommend using the NCBI Taxonomy Browser to identify their parent groups.

Q: I received an error when uploading a list of species.

A: Make sure that you are uploading a TEXT file (.txt) with one taxon per line. These taxa should be in scientific nomenclature as per the NCBI Taxonomy Browser. If your file is correct and you are still receiving issues, it could be that there is heavy usage of our servers. Please try again later and email [email protected] if you are still experiencing issues.

Q: Why are some of the taxa in the prune tree search marked with an asterisk (*)?

A: When a taxon does not appear on the TimeTree guide tree, TimeTree may be able to infer its position from one of its parent taxa, and provide a topologically equivalent substitute. The reliability of these inferences are dependent on the accuracy of the NCBI Taxonomy. The identity of the substituted taxa can be found by clicking on the individual nodes to open the Node Summary. Failed substitutions will be listed under Unresolved Names.

Q: How is topological incongruence among studies resolved?

A: First, we use divergence time estimates to resolve the TTOL rather than topological data from studies. Then we evaluate these resolutions for topological concordance with studies. In cases where resolutions are discordant with a majority of studies, we test alternate local configurations, adopting those that minimize discordance. Many nodes in the tree of life are poorly resolved and the robustness of this solution is limited by existing data. We recommend that researchers consult the most recent phylogenetic literature, some of which is not in TimeTree because not all phylogenetic studies include a dated phylogeny. For more details about the acquisition of data for TimeTree, standardization, and the assembly of a global timetree (TimeTree of Life, TTOL) based on individual timetrees, please see Kumar et al. (2022) and Hedges et al. (2015).

Q: How did you derive confidence intervals for molecular time estimates?

A: Our primary aim is to inform users about differences that exist among studies in estimated time through the presentation of "confidence intervals." We base our confidence interval on the Empirical Rule (95%) in statistics, which states that, assuming a normal distribution, approximately 95% of the times reported across studies will fall within two standard deviations of the mean time. That is, we use the 95% Empirical Rule for a population of studies, which has an interpretation similar to confidence intervals and hypothesis testing. Intuitively, we expect that the among-study variance, obtained based on study node times as single data points, captures a variety of lower-level errors including differences in calibrations and sampling of genes and taxa. We present this confidence interval for nodes where times are available from five or more studies - otherwise a min-max range of time estimates is given (or no range if only a single estimate is available). We provide The Empirical Rule confidence interval as a quick guide to the amount of variation among studies, and we strongly recommended that researchers review individual studies and their methodologies before using any time estimates in downstream research.

Q: What search method is employed to locate molecular time estimates for a pair of taxa?

A: When users submit a valid pair of taxa through the search page, TimeTree determines the most inclusive taxonomic groups for each pair of query taxa. Such groups are one level lower in the taxonomic classification than the common taxon that includes both query taxa. Once these groups are determined, TimeTree obtains all molecular time estimates in which a member of the most inclusive group for Taxon A is compared to a member of the most inclusive group for Taxon B.

Q: Why are the calibrations not listed for each study?

A: Calibrations are important in timing but different investigators use them in different ways. For example, they can be used as minimums, maximums, fixed points, and with different distributions (e.g., lognormal, exponential, uniform). Two studies with identical calibrations could present greatly different time estimates because of this complexity in methodology. If you have questions about such details, please refer to the original studies.

Q: Why is the time estimation method not listed for each study?

A: There are many methods of time estimation and many variables within each method. Two studies with identical data, calibrations, and using the same software, can yield different time estimates depending on how the software is used. If you have questions about such details, please refer to the original studies.

Q: How can I download the data?

A:

1. To download a Newick file of a timetree displayed in a "Build a Timetree" search, use the option on the bottom left column, "Export Tree" and select "To Newick File."
2. To download an individual study timetree, click on the "Studies" button in the toolbar, find the study of interest, and click on the "tree" link at the right side of the page. The timetree will be displayed, and then you may export the Newick file using the button under "Export Tree." Individual study trees (Newick format) are also accessible directly from Timetables.
3. To download all individual study timetrees, please send an email request to [email protected].
4. To download the complete timetree of 142,103 species, please send an email request to [email protected].
5. Data can be downloaded through individual API requests as described in Kumar et al. (2022)..

Q: How often is TimeTree updated?

A: Currently, TimeTree is updated periodically, because each update requires changes to the large guide tree. We are moving towards regular and more frequent updates as our technology improves.

Q: Can I submit my own, unpublished, molecular time estimate to TimeTree?

A: TimeTree currently relies solely on published data and does not accept unpublished time estimates.

Q: How do researchers use TimeTree in their work?

A: TimeTree has been cited in hundreds of studies across a broad spectrum of sciences - from obvious topics such as genomics and evolution to more niche fields such as paleovirology and behavioral transcriptomics. TimeTree is used for calculating divergence times, calibration points for building new timetrees, phylogenetic topology, and complete timetrees. We encourage you to follow our social media accounts where we regularly share studies that use TimeTree in a variety of ways. See our Twitter and Instagram pages.

Q: Can I get a list of the literature referenced by TimeTree?

A: Yes, please refer to the Studies page.

Q: Can I download the software used to build the TimeTree of Life?

A: Source code for software used to construct and display TTOL v5.1 is located at this Github page.

Q: I found an error on the TimeTree website. How can I report it?

A: Please submit any errors or suggestions to us.

Q: What previous versions have been published?

A: You are viewing TimeTree v5.1 (2023) with 4,185 studies and 150,635 taxon tips (148,876 species). Previous versions released include version 1 released in 2006 with 70 studies, version 2 released in 2011 with 910 studies and 17,000 species, version 3 in 2015 with 2,274 studies and 51,000 species, version 4 in 2017 with 3,000 studies and 97,000 species, and version 5 in 2021 with 4,095 studies and 142,103 species.