Of course you can map your RNA-seq reads to relative genome. But, I don't know why you would like to do so. If you would like to determine expression level of each transcript, you could align those reads back to their assembled transcriptome.

You should map your reads to your assembled genome and then perform a BLASTp of your non-model organism predicted proteins to Tribolium castaneum (and maybe D.melanogaster...) in order to identify conserved genes / proteins of your non-model organism with other organisms, and thus infer their potential functions.

Before mapping your reads to another organism genome, you should compare the genomes of the two organisms to verify how close they are.

To solve the annotation problem, you can adopt various paths. 

1. You can use the RNAseq reads to predict gene models on your assembled genome (Tophat2+Cufflinks) and quantify their expression at the same time. 

2. You can use the genome of a closely related organism and transfer their annotation. (gene models) See http://ratt.sourceforge.net/ for a very good tool that does just that!

With any of these options, one should be always careful of assessing the predicted function of the gene models and to add some level of functional annotation, as Gautier Richard mentioned. There are a number of tools to use for that.

A. OrthoMCL for grouping of genes into orthologus groups - you will need annotations from other organisms to do this (I think)

B. Use Pfam to predict conserved proteins domains (highly recommended)

C. Interproscan - will also assign GO terms to your annotations - very useful

Good luck

Hi @Kun Lu, @Gautier Richard and @Anna Protasio.

Yes I already did DEG analysis by mapping the reads to assembled sequences. The reason why I want to align this MP( an insect) transcriptome data  to Triblium castaneum genome is for the following reasons:

First, my sample insect (MP) and Triblium castaneum is closely related (they blong to the same family).

Second, the environment they live in are same (Triblium castaneum is in flour; MP in the desert).

So by mapping MP transcriptomic data to Triblium castaneum genome I want to try to find common genomic information among these two species, and then to screen some drought resistant genetic information too.

I have downloaded Triblium castaneum genome, all the transcript and protein sequences from NCBI.

And what I want to do is mapping MP transcriptome data (reads or assembled sequences) to Triblium castaneum genome, and find out which unigene mapped to gens in the genome, and the gene’s position.

I am so new at this, I do not know how go about it first.

Hi Kurban,

Your most recent post places different questions to the original one. It might be helpful for you to start a new thread with a more informative title.

In any case, it seems to me that your question is not just about mapping reads/transcripts to a genome but more regarding comparative genomics. If you want to "find common genomic information" between your genomes of interests, it might be useful to check out genome to genome comparison tools - Artemis and ACT from the Sanger Institute provide useful tools for visualisation of genomes (note that you will have to do blast these genomes to find the most conserved regions. 

You can then find the resistant traits (coding sequences. SNP) and visualise these regions in the context of the genomes. You can overlay information of gene models (maybe from your cufflinks output). 

Anna

http://www.sanger.ac.uk/science/tools/artemis

http://www.sanger.ac.uk/science/tools/artemis-comparison-tool-act

Hi Anna,

Maybe I did not express myself clear, sorry about that. What we have is transcriptome data (sequenced reads and assembled sequences), because our sample is non-model orgasm. We want to align it to Triblium castaneum genome (model species).

I doubt that mapping your non-model organism read/transcriptome to the model-organism genome assembly would help you answer your question (which I understand it is still "find common genomic information").

But if aligning is what you want to do you should give it a go. How informative these results are with respect to your question will depend on how closely related your organisms are. Something along this lines is discussed in the paper - see link.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3178736/

Read mapping is very sensitive to differences between source genome and reference - which is why SNP calling is even possible.

Mapping reads to a different species (even a close one) is unlikely to give you what you're after; too many reads won't map at all to the genome.

I agree with Anna and Gautier you need to compare the genes/proteins of the two species to be able to infer functional information. The RNA-seq reads are best used to identify which TM genes are important to you and then identify homologues in TC via BLAST, PSI-BLAST or HMMer.

Hi Kurban,

Nucmer (http://mummer.sourceforge.net/manual/) can give you very quick alignments of your assembled transcript to the T. castaneum genome, if the nucleotide sequence similarity is high enough. If your organism is too different from T. castaneum to allow good nucleotide alignments, Promer from the same source may be useful. I haven't used Promer myself.

Scipio (http://www.webscipio.org/) is a good tool for precise alignments of proteins to genomes.

The Blast methods suggested by others here are also good for identifying homologous genes / proteins in a related species.

When you are working with a little-studied or non-model organism, trying several methods and comparing their results is worthwhile to find which methods give useful results for your organism and to build confidence in your conclusions.

Happy experimenting!

For aligning raw sequence reads to the reference Triblium castaneum genome, you may try TopHat (https://ccb.jhu.edu/software/tophat/index.shtml) that is integrated with the ultrafast aligner Bowtie (http://bowtie-bio.sourceforge.net/index.shtml). For aligning de novo transcript assemblies to the reference genome, why not trying BLAT (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC187518/) or GMAP (http://m.bioinformatics.oxfordjournals.org/content/21/9/1859.long)?

For RNA-seq data it is recommended to use a spliced aligner like STAR, which is one of the best and most practical aligner to date (followed by Htseq count or featureCounts for counting reads per gene before Differential Gene Expression analysis).

Bowtie2 is used to align DNA-seq data.

TopHat is generally considered to be OK, even if it's not the most practical aligner, but avoid using Cufflinks for DEG analysis / counting / normalization, the algorithms used are kind of a black box and generally perform poorly compared to others. It is recommended to use EdgeR and Limma on raw reads counts assessed as described above (where normalization in CPM or RPKM / FPKM is performed within R).