3.8 years ago by
Spain. Universidad de Córdoba
It is likely we need some more details..
I would expect you to run first an alignment (a mapping) with your fastq files. You can include the replicates in that alignment by including all of the corresponding fastq files in the command lane
That will provide you with a unique BAM file where you can run your htseq-count program
I recommend you to give a look to the new pseudo aligners such as Salmon or kallisto in combination with processing R programs like sleuth if you are planning to do a DE analysis of your data and are not interested initially in doing an isoform analysis of your reads
Kallisto will do the alignment in a question of minutes if you provide a multi fasta file with the cds of interest. I believe you are working with Arabidopsis, and that fasta cds file is already provided in the TAR database.
Kallisto needs a very short and easy command which can be 1/100th less complicated that one used for TopHat or another mapping software (even using Galaxy). Sleuth needs only less than 10 codes lanes to give your answers for DE expression. And everything correlates very well with data obtained through "traditional methods". A search in the bibliography will compare the advantages and disadvantages of the two
You can also use kallisto and DESeq2 if you want. In the DESeq2 vignette you will find information on how to accomplish this