**RNA-seq Library Construction**

Based on histological determination of gender and developmental stage, seven female and six male paraffin-embedded gonad samples were selected for RNA isolation. The female samples represented three developmental stages including; stage 1 (No secondary ooctyes; or secondary oocytes ~ 5 - 15 µ) n=1, stage 5 (Secondary oocytes ~50 - 75 µ) n=3, and stage 6 (Secondary oocytes ~65 - 85 µ) n=3.  Males samples represented two developmental stages ;  stage 1 (~ <5% spermatids per acinus; small acini) n=3 and stage 3 (~50% spermatids per acinus) n=3.  RNA was isolated using the PAXgene Tissue RNA Kit (Qiagen) according to the manufacturer’s instructions with the following changes. Five 5um sections were collected from each paraffin block. Where the protocol indicated centrifugation speeds of “maximum speed”, samples were centrifuged at 19,000g. In lieu of a shaker-incubator, the samples were incubated for 15 minutes at 45C in a dry oven using the Disruptor Genie (Scientific Industries). Samples were eluted using 40uL of Buffer TR4. 

All samples were quantified using a NanoDrop1000 (Thermo Fisher Scientific). RNA integrity was verified on a Bioanalyzer (Agilent Technologies) using either the RNA 6000 Pico Kit (Agilent Technologies) or the RNA 6000 Nano Kit (Agilent Technologies), depending on sample concentration.

Two pools of RNA (male RNA and female RNA) were prepared for RNA-seq. Both pools contained equal quantities of RNA from each developmental stage. Library preparations and sequencing were performed by GeneWiz, Inc. Libraries were constructed using the Ultra RNA Library Preparation Kit (New England Biolabs). Both libraries were sequenced on a single lane of the HiSeq2500 (Illumina) sequencer.

**Sequence Assembly**

Raw sequence reads were quality trimmed using trim galore (citation) [code/output](https://github.com/sr320/paper-pano-go/wiki/Quality-trim-output), and quality assessed using FastQC (citation). Assembly of sequence reads was carried out using Trinity (citation) using the following parameters...

Change to...
Raw sequence reads were quality trimmed using trim galore (Krueger 2013), and quality assessed using FastQC (Andrews 2010) eliminating adapters and with a 20 quality Phred score cutoff (Ewing et al. 1998[not sure about this reference]). Assembly of sequence reads was carried out using Trinity (Grabherr et al. 2011) with default parameters.