Molosid bats

Sensitivity of partitions to phylogenetic reconstruction by maximum likelihood(ML) and Bayesian inference(IB) in molosid bats (Molossidae Family, Molossinae subfamily).

Introduction

The molossidae family contains more than 100 species divided into 16 genera, positioning it as the 4 largest family of bats. Two subfamilies are recognized, Molossinae (15 genera) and Tomopeatinae (1 genus, with only one species: Tomopeas ravus.). The main characteristic of the molosidos is the presence of free tail, measured the forearm between 25 to 86 mm, extension of the ear, shape of the antitrago, wrinkles on the lips and the presence of bristles on the rump. As diagnostic characters, the Molossinae subfamily presents a thick uropatagio, The tail usually extends a third or more of its total length beyond the distal edge of the interfemural membrane, The fur is velvety, the ears are well developed, thick and provided of a prominent internal keel in most species, in addition to the presence of tiny, rigid and curved hairs, with the apex dilated in the external toes of the feet and long sensitive hairs starting from the medial region of the feet (Gregorin 2001) .

Phylogenetic relationships within Molossidae have been investigated using morphometry (Freeman, 1981), morphological characters (Legendre, 1984, 1985, Hand, 1990), and molecular data (Sudman et al., 1994). The phylogenies derived from these studies are largely incongruent, but most propose Molossinae as monophyletic. That is why the objective of this study is to determine the effect of partitions in phylogenetic reconstruction by ML and IB in the Molossinae subfamily.Taking as a hypothesis that IB is better than ML, given the support of the nodes of the tree.

Methods

GenBank sequences were downloaded for three nuclear genes (RAG2, DMP1 and βFIB) and one mitochondrial gene (ND1) (Annex 1), for 38 species of which 36 represent the ingroup at least one species for each genus and 2 the outgroup ( Myotis velifer and Antrozous pallidus of the family Vespertilionidae). The sequences were aligned in Muscle v3.8.31. Evolutionary models of each gene were calculated with jModelTest v2.1.10 under the Akaike Information Criterion (AIC).

Two methods of phylogenetic reconstruction were used: Maximum Likelihood (ML) and Bayesian Inference (BI). For ML, RaxML v8.2.9 (Combining tree search and bootstrapping analysis) was used. For IB the software of MrBayes v3.2.6 was used. For the two methods, 3 types of data were used: Nuclear, mitochondrial and total evidence (nuclear + mitochondrial).

Results and Discussion

The substitution models were: RAG2: TIM1 + G, DMP1: TrN + G, βFIB: TPM2uf + I and ND1: GTR + G + I.

 The trees obtained with the two methods for mitochondrial and nuclear data separately presented little resolution of the phylogenetic relationships within the molossinae subfamily; On the contrary, for total evidence in the two methods, trees with higher resolution are obtained but the relationships within the subfamily are not completely resolved. Some genera such as Mops, Chaerephon and Tadarida in none of the cases were found as monolithic groups.

The bootstrap support for the ML tree is very low for some nodes (Image 1), whereas the posterior probability for most of the nodes of the tree obtained by IB is 100% (Image 2).

These results are similar to that found in the literature about this subfamily; Lamb et al. (2011) in the non-monophyly of the genera Mops, Chaerehon and Tadarida, in the case of Tadaria, Gregorin, 2015 also supports non-monophyly. Like the clade Chaerephon-Mops, this pattern was also recovered by Gregorin (2000) and Freeman (1981).

On the other hand, as the objective of this work was to determine the effect of partitions on phylogenetic reconstruction by ML and IB in the Molossinae subfamily; I can affirm that the approximations of only mitochondrial or nuclear data coincide with those found by other authors for the phylogenetic relationships of this subfamily (Gregorin (2000) and by Freeman (1981) Lamb et al. (2011)), besides the idea is supported of using total evidence for phylogenetic analysis. Regarding the methodology, the results are similar, but as you can see there are long branches that may be affecting, but according to the literature, Bayesian inference presents consistency regardless of the length of the area or branch (Steel, 2013), contrary to the ML, for which the most reliable result is obtained by IB.

Conclusions

The phylogenetic reconstruction of the subfamily is affected by the type of data with which the analysis is performed, being more appropriate an approach with total evidence by Bayesian Inference.

Image 1. Tree obtained from ML with bootstrap supports.

Image 2. Tree obtained by IB with posterior probability values.

Bibliography

Darriba, D., Taboada, G. L., Doallo, R., & Posada, D. (2012). jModelTest 2: more models, new heuristics and parallel computing. Nature Methods, 9(8), 772–772. http://doi.org/10.1038/nmeth.2109

Freeman, P. W. 1981. A multivariate study of the family Molossidae (Mammalia, Chiroptera): morphology, ecology, evolution. Fieldiana Zoology, 7: 173 pp.

Gregorin , R. 2000. Filogenia de Molossidae Gervais, 1856 (Mammalia: Chiroptera). Ph.D. dissertation, Universidade de São Paulo, São Paulo, Brazil.

Gregorin, Renato &Cirranello, Andrea.2015. Phylogeny of Molossidae Gervais (Mammalia: Chiroptera) inferred by morphological data. Cladistics(31).doi:10.1111/cla.12117

Lamb,J; Ralph,T; Theshnie Naidoo, Peter J. Taylor, Fanja Ratrimomanarivo, William T. Stanley, and Steven M. Goodman.2011.Toward a Molecular Phylogeny for the Molossidae (Chiroptera) of the Afro-Malagasy Region Acta Chiropterologica 2011 13 (1), 1-16

Legendre, S. 1984. Etude odontologique des representants actuels du group Tadarida (Chiroptera, Molossidae). Implications phylogenetques, systematiques et zoogeographiques. Revue de Suisse Zoology, 91:99-442.

     Legendre, S. 1985. Molossides (Mammalia, Chiroptera) cenezoiques de l'Ancien et du Nouveau Monde; statut systematique; integration phylogenique des donnees. N. Jahrb. Geol. Palaont. Abh., 170: 205-227.

Hand, S. J. 1990.First tertiary molossid (Microchiroptera: Molossidae) from Australia: its phylogenetic and biogeographic implications. Mem. Queensland. Museum, 28:175-192.

Stamatakis, A. (2014). RAxML version 8: A tool for phylogenetic analysis and post- analysis of large phylogenies. Bioinformatics, 30(9), 1312–1313.http://doi.org/10.1093/bioinformatics/btu033

Steel, M. (2013). Consistency of Bayesian inference of resolved phylogenetic trees. Journal of theoretical biology 336: 246–49.

Sudman, P. D., L. J. Barkley, and M. S. Hafer, 1994. Familial affinity of Tompoeas ravus (Chiroptera) based on protein electophoretic and cytochrome B sequence data. Journal of Mammalogy 75:365-377.

Robert, E. C. (2004). MUSCLE: Multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Research, 32(5), 1792–1797. http://doi.org/10.1093/nar/gkh340

Ronquist, F., Teslenko, M., Van Der Mark, P., Ayres, D. L., Darling, A., Höhna, S., Huelsenbeck, J. P. (2012). Mrbayes 3.2: Efficient bayesian phylogenetic inference and model choice across a large model space. Systematic Biology, 61(3), 539–542. http://doi.org/10.1093/sysbio/sys029