# FAQs {% if data.FileExists("Filter") %} ## Data filtering ### What are the requirements for sequencing data quality? Q20 and Q30, generally we require Q20>=90%, Q30>=85% ### Why does the base quality distribution of Clean Reads appear in the form of a horizontal line,not a point? Are these lines based on the average value? Is it discontinuous data? 1. In fact, the points are too dense and connected into a line; 2. This is the base quality of the corresponding position on each read, which is the actual value, rather than the average value; 3. It is indeed discontinuous, However, regardless of the position or quality, the digital interval is integer 1. The data is PE90 sequencing, so at position 90, high-quality bases will be broken, which is caused by the poor 5'stability of the sequencing. {% endif %} {% if data.FileExists("Quantification") %} ## Quantification Analysis ### What is the difference between FPKM and RPKM? The calculation methods of FPKM and RPKM are basically the same. The difference is that FPKM calculates fragments, while RPKM calculates reads. Fragment has a broader meaning than read, so FPKM has a broader meaning,either a of pair-end fragment or a read. RPKM was born for SE sequencing, and FPKM is a correction for RPKM on PE sequencing. ### Is biological repetition needed? How many times? Yes, at least two biological repetitions are required, and more than three biological repetitions are better. An article published by Hansen [1] in July 2011 showed that biological differences are the characteristics of gene expression, and is irrelevant with detection technology and data processing methods. If there is no biological duplication, the paper may be rejected by magazines with high impact factors. ### What is the ideal correlation coefficient between samples? Based on the criteria recommended by the Encode program, the square of the correlation Person coefficient (R2) of two sample replicates should be ≥ 0.92 (ideal sampling and experimental conditions). Reference: Standards, Guidelines and Best Practices for RNA-Seq https://genome.ucsc.edu/ENCODE/protocols/dataStandards/ENCODE_RNAseq_Standards_V1.0.pdf {% endif %} {% if data.FileExists("Coverage") %} ## Gene Coverage Analysis ### Can gene coverage be accurate to the transcript level? Now can't achieve, because we perform statistics based on the mapping results. When there is overlap between transcripts, we cannot distinguish between transcripts but give the coverage of the transcript interval. Therefore, when need to use the transcript coverage, we must consider its expression level. Because RSEM uses the maximum likelihood method to distinguish the read allocation of different transcripts when calculating the expression level, but it does not give the details of the assignment, only the expression level. {% endif %} {% if data.FileExists("Cluster_Mfuzz") %} ## Time series analysis ### What is the difference between time series analysis and general clustering methods? Time series analysis uses relative amounts for clustering, so it is very sensitive to changing trends. The general clustering method uses absolute amounts for clustering, so it is more likely to cluster genes with similar expression level together. ```image url: {{data.ReportRoot}}/md/pic/cluster.png subTitle: Figure 1 The difference between time series analysis and general cluster analysis ``` {% endif %} {% if data.FileExists("Rich") or data.FileExists("Rich_target") or data.FileExists("Rich_mfuzz") or data.FileExists("Rich_wgcna") %} ## GO Enrichment Analysis ### There are three prefix "go_c", "go_f", and "go_p" of the result files. What do they mean? They are the three major categories of GO: C cellular component, F molecular function, and P biological process. {% endif %} {% if data.FileExists("Rich") or data.FileExists("Rich_target") or data.FileExists("Rich_mfuzz") or data.FileExists("Rich_wgcna") %} ## KEGG Enrichment Analysis ### Why does red and green color appear in one of the boxes the same time of the KEGG's pathway diagram? In the box which are some of the enzyme or the protein with catalytic activity basically.And we know that some of the key enzymes or important proteins are encoded by multiple homologous genes.And some isoenzyme are aggregated from multiple subunits, each subunit is encoded by a gene.That is, these important enzymes or proteins encoded by different genes (transcript), then sometimes there will be inconsistencies in gene expression. For this part of the results, it is recommended to skip the look. {% endif %} {% if data.FileExists("SnpIndel/SNP") or data.FileExists("SnpIndel/Indel") %} ## SNP and InDel Analysis ### What is the format of VCF file? The VCF format is used to store variant data information, such as SNP, INDEL, CNV and SV, etc. For details, please refer to the official description of UCSC http://genome.ucsc.edu/FAQ/FAQformat.html#format10.1 {% endif %} {% if data.ProjectInfo.ProjectType == 'lncRNA' or data.ProjectInfo.ProjectType == 'Hiseq_lncRNA' %} ## miRNA Precursors In lncRNA Prediction ### Will the lncRNA only be mapped to miRNA precursors of this species? We mapped lncRNA to the entire mirbase database without distinguishing between species. Because the alignment is performed independently, if you only focus on the mirRNA precursors of this species, you can ignore the results of other species. ## lncRNA Family Analysis ### How to find detailed information about a gene family? The gene family's database is derived from xfam, and you can find detailed family description information on its official website. Official website: http://rfam.xfam.org/ {% endif %} {% if data.FileExists("Target")%} ## lncRNA Target Gene Prediction ### What is the mechanism of Cis regulation? The mechanism of cis regulation is very complex. could through impacting the polymerase regulation, by affecting the promoter, etc. Please refer to these two reviews: 1) Gene regulation by the act of long non-coding RNA transcription: https://www.ncbi.nlm.nih.gov/pubmed//?term=23721193; 2) Regulatory mechanisms of long noncoding RNAs in vertebrate central nervous system development and function ：https://www.ncbi.nlm.nih.gov/pubmed//?term=23337534 {% endif %} ### How do we pick the verified gene in the qPCR experiment? Generally it is necessary to verify lncRNA and its target gene. It depends on the original intention of sequencing, or the biological problem you want to solve, and then combine the results of GO enrichment and KEGG enrichment to see if there is any expression difference in the gene you are concerned about. It is suggested that the genes with high expression and high fold change of genes expression and high coverage should be selected. When a gene with a low expression is selected for qPCR validation, the probability of not meeting the sequencing results will be increased.