Dear All
I came across the paper below and would like to apply the methods for my study.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5855742/
De Novo Assembly: High-quality clean reads were de novo assembled using Trinity v2.0.6 software without a reference genome. The Trinity assembly results were called transcripts,and then gene family clustering was performed using TGICL clustering software v2.0.6 to obtain the final unigenes. The unigenes were divided into two classes:clusters (CL), composed of several unigenes with a shared similarity greater than 70%, and singletons (unigenes).
Differentially Expressed Unigenes (DEGs): The clean reads from each sample were compared with the reference unigene database using Bowtie2 v2.1.0 software. The unigene expression levels were then estimated using RSEM v1.2.12 with the default parameters for each sample, employing FPKM values. DEGs between groups (XF versus XM, XM versus XN, and XF versus XN) were analyzed using the PoissonDis algorithm, which is based on the Poisson distribution, performed as described by Audic. For screening purposes, an FDR ≤0.001 and a fold change ≥2.00 were used as the thresholds for identifying significant DEGs
From my understanding, the author first performed de novo assembly using Trinity. Then, the author proceed to obtain unigenes using TGICL. Next, RSEM was run to obtain FPKM values.
However, I could not understand how they determined the DEGs. "DEGs between groups (XF versus XM, XM versus XN, and XF versus XN) were analyzed using the PoissonDis algorithm, which is based on the Poisson distribution, performed as described by Audic. For screening purposes, an FDR ≤0.001 and a fold change ≥2.00 were used as the thresholds for identifying significant DEGs".
- May I know which software can run PoissonDis algorithm?
- Is that they run PoissonDis algorithm first before proceed with screening using FDR and fold change or all these can be run together?
I'm asking this as I could not find more details to analyze Differentially Expressed Unigenes (DEGs) using Unigenes generated from TGICL. Most of the study directly used the assembled data from Trinity to performed gene-level differential expression analysis. However, this method does not suit my current study objective which is study the gene function by looking at the gene level.
Hopefully someone carry out the similar study can guide me through. Thanks!
