hello... m working on ngs. i have 4 samples and i got the coding gene information too. all the four samples have around 2800- 2900 genes now i have to look for the common genes found in all the four samples. but i have no idea about it. friends can u help plz ?
•
written
by
Arpitha Hebbar
Asuming files with one gene name per line
cat file1 file2 file3 file4 | sort | uniq -c > gene _sample.counts
gene_sample.couts will contain how many times each gene is present out of four samples.
If you have gene info in bed format then u can use bedtools to get intersection of them.
if inputs are fasta sequence files : u can cd-hit-est to cluster sequences at specific seq identity and with liitle scripting u can get get how many in cluster comming from different samples.... you even try uclust..
m working on ngs data. i dnt know the name of the genes. i know only the number of genes. this i got from the prodigal server. so how can i found out ?
data generated from next generation sequencing was assembled and m having the contigs which is in the fasta format. by using the prodigal server i calculated the number of genes. but i dont know their names. i wanted to find out the common genes found in all the four samples. each having the 2871,2828,2983,2895 genes.
okay... thnx ! ll try but if it uses the gene name to distingwishes then i cant use because all the gene names are names as gene 1,gene 2, gene 3....... and so on
Log in to answer this question.
Could you please expand a little bit on that?
In your "coding gene information" files, do you have annotation from the same source, so that you only want to compare the names of the genes in all four files? Or do you have roughly similar sequences and want to know which software to use that aligns these sequences and tells you the shared areas?
i am doing the ngs analysis. i have four samples contig file. i got the totla number of coding genes from prodigal software. even i dont know the name of the genes. so how can i find out the common genes in four samples.
What do you mean with "common" genes? Do you mean orthologous genes? If it is so, you can use Orthomcl or Proteinortho to find them. You'll need the fasta files of all the proteins
m having the fasta files and mine is orthologous sequnce ! i ll try thanku...!