Description of the data and file structure

This repository is associated with Abondano Almeida, D., Anganoy-Criollo, M., Grant, T., Klimovych, S., & Schulte, LM (2025). Histological and gene expression analyses of the arm and finger macroglands of two Hyloxalus frogs (Dendrobatidae). Molecular Ecology, 34(22), e70162. https://doi.org/10.1111/mec.70162
We studied the integument of the arms and fingers of the Hyloxalus nexipus and H. azureiventris males, focusing on glandular composition, gene expression patterns, and potential pheromone production. For this, we used a whole-transcriptome sequencing (RNAseq) approach and histology analysis. For RNA sequencing, we followed the methods and pipeline of Abondano-Almeida et al. (2024). Our RNA-seq results provide evidence of the co-occurrence of two sexually dimorphic macroglands—swollen fingers and BAG—in H. nexipus, which is a rare trait in dendrobatids. Both structures differentially expressed SPF, suggesting complementary roles in courtship. In H. azureiventris, however, no differential SPF expression was found in the arm, leaving its reproductive role uncertain. Both species expressed SPF in their fingers. Thus, based on other dendrobatids where SPF is upregulated in male fingers with SMGs linked to pheromone production, we hypothesize that H. azureiventris may have specialized glands despite lacking visible swelling. Our findings reveal a novel pheromone-producing gland and underscore the complexity of chemical communication in dendrobatid reproduction. 
This repository contains the assembliesthath are necessary for performing annotation and the differential expression (DE) analyses. In turn, the DE analysis was the input for the volcano plots, graphs that show the gene expression in the two conditions (glandular tissue vs. non-glandular tissue). The DE analysis was conducted using the R package DESeq2 (version 4.2). The raw sequencing reads necessary for all gene expression analyses are available from the European Nucleotide Archive under project accession number PRJEB102072.

Files:

• Hyloxalus.azureiventris_arm-leg_assembly.tr

• Hyloxalus.azureiventris_finger-toe_assembly.tr

• Hyloxalus.nexipus_arm-leg_assembly.tr

• Hyloxalus.nexipus_fingers-toe_assembly.tr

Description: The assemblies are presented in fasta files (.tr). Those files contain the final assembly per species from all males per tissue type (glandular tissue or non-glandular tissue). For each individual, we used kmer sizes of 25, 53, and 75 in SPAdes v.3.12 (Bankevich et al. 2012), incorporating both sets of reads (arm and leg skin; fingers and toes). With four individuals sequenced per species, this resulted in a total of 12 assemblies for each tissue type per species. For H. nexipus, only 11 assemblies were produced for finger and toe samples due to computational limitations preventing the use of kmer 25.

TABLE_S11.xlsx
Description: The list of annotated genes and differential expression analysis from both species, H. nexipus and H. azureiventris, per tissue type (arm-leg or finger-toes) is presented in an Excel table. For each species and tissue type (in different tabs in the Excel file), we annotated the final merged (evigene) assembly (using Diamond v.0.9.22.123; Buchfink et al., 2015), using Blastx mode to search assembly transcripts against the Identical Protein Groups database from NCBI (NCBI, 2022), which was filtered to include only vertebrate sequences. Diamond was run with max-target-seqs 1 and max-hsps 1 in order to retain a single, best hit for each transcript. The differential expression (DE) analysis was performed with the R package “DESeq2” (version 4.2). The Benjamini-Hochberg correction was applied to p-values to account for multiple hypothesis testing (Love et al., 2014). Given that we employed a paired-sample design (gland and control tissues taken from each individual), the final model included tissue type as a fixed effect and individual as a random effect to account for both treatment and individual variation. The dataset has a header, and this file can be opened with any text editing software.
3.1 Tabs:
H.azureiventris.arm-leg
H.azureiventris.finger-toe
H.nexipus.arm-leg
H.nexipus.finger-toe
3.2 Variables-Header
• id_transcript (Column A): Refers to the NCBI code for a coding protein (annotation) given to the transcript. Thus, this value is not unique and can be present several times in the column
• transcript (Column B): a unique value for every transcript in the final assembly

• annotation (Column C): NCBI code and name for the protein coding gene

• baseMean (Column D): metrics from the DE analysis

• log2FoldChange (Column E): metrics from the DE analysis. Quantifies the magnitude and direction of gene expression change between experimental conditions

• lfcSE (Column F): metrics from the DE analysis

• stat (Column G): metrics from the DE analysis

• pvalue (Column H): metrics from the DE analysis

• padj (Column I): metrics from the DE analysis. The adjusted p-value, is the crucial output for identifying significant differentially expressed genes after correcting for the massive number of tests performed (multiple testing problem) using the Benjamini-Hochberg (BH) method to control the False Discovery Rate (FDR)

• TPM_1arm (Columns J-Q): Transcript per Million (TPM) from the individual one (1) from the arm. In each consecutive column (e.g., column K= TPM_2arm), the TPM is followed by the individual number and the tissue type.

REFERENCES Abondano Almeida, D., Twomey, E., Vargas-Salinas, F., Meyer, C. & Schulte, L.M. (2024) Sexy fingers: Pheromones in the glands of male dendrobatid frogs. Molecular Ecology, 33: e17476. https://doi.org/10.1111/mec.17476