Many to Many alignment of Large number of sequences
Hello, Is there some tool to align many to many sequences for a large number of sequences? I have around a set of 10 million sequences to be aligned to another set of 20,000 sequences. I would just like to have an output of number matched in 20,000. Suggestions would be really helpful. Thank You very much
• 2,344 views
•
link
updated
by
Biostar
•
written
by
bharath.mh24 •
•
written
by
bharath.mh24 •
0 answers
No answers yet.
Log in to answer this question.
More posts like this
-
NCBI MEGABLAST versus UCSC BLAT
written by a615ebfb •
-
BWA-mem - Different alignment numbers to same reference sequence
written by Steven •
-
Comparing fastq files
written by diffexp123 •
-
protein-protein interaction network
written by zelda •
-
Alignment-free counting of sequence abundance
written by Saima •
-
Finding % conserved proteins
written by max_19
-
Multiple alignment: compare one sequence with many
written by mdihcroe •
-
Mapping reads to Primers Combinations(Custom Reference Genome)
written by bharath.mh24 •
-
Summarizing the Results in R
written by alizohaib7 •
-
Order Sequences By Number Of Times Found
written by Pauln •
You can probably use EMBOSS for this. However, I am wondering how you want to interpret your results (10M with 20k alignments).
Yes. I thought of needleall. But does it take so many sequences as Input? Also, i would just want to know the number of times a sequence matches to the 20,000 sequences. Kind of like mapping!
This is how I understand your problem set.
You have a set of 20000 sequences, for the purpose of solving this task, lets call it your reference sequences.
You have a set of 10 million sequences and that is your candidate sequences.
Based on the computational sources at your hand, build a local blast database, following the instructions here:
https://www.ncbi.nlm.nih.gov/books/NBK279688/
Then blast all your 10 million sequences to match against this blast database. I hope this is a right direction to solve your problem.
Will I be able to blast so many sequences at once? Is there some way I could make use of multiple cores or speeden it up? Because I tried this and it is always stuck. Thank You
What kind of computing resources do you have at your disposal?
Server offering computing cluster with processor Intel Xeon E5-2698v3 16C/32T 2.30 Ghz, 63 nodes and 32 cores. Computing resources dont seem to be the problem. I would think I need to change the command a bit from just blastn. Thank You Very Much
You may be better off debugging why you are not able to make bowtie or bwa work (I will suggest that you try bbmap). Is the data trimmed/cleaned (adapters etc)?
Dealing with/summarizing blast output from that massive search would present a challenge of a different kind. Plus you would not have standard SAM/BAM files that downstream tools expect.
I did try working on debugging that for a while. But I couldn't quite figure anything out! Could not think of anything! Thank You
Unless you post command lines you tried/errors you have encountered we can't help.
Assuming you mean pairwise local alignments, I would use BLAST for this.
Yes, but then the output format would be too large to access. I am not very skilled in programming, so any suggestions would be helpful!!
You can use BLAST (or DIAMOND) with outfmt 6 (tabular) and then count the number of hits with sort & uniq, e.g.: cut -f 2 output.txt |sort |uniq -c
Oh, Okay. I shall try it! Thank You very much!!
But how do you blast one set of sequences with another set? Basically, a many to many blast.
Two multi-fasta files, one is the query, the other is the database (should format it with makeblastdb first) and then run one against the other
Before we all recommend the wrong tools: are your 10 million sequences sequencing reads (and maybe even of the same/very similar length), by chance?
Yes, it is! 5C but I have been recommended to align the reads!! Thank You
From your description it sounds like you were given a set of NGS data, like RNA-seq, ChIP-seq, BS-sew and a set of scaffolds for an assembly of a plant or animal. Could you please tell us what is dataset tipe you were given? This will affect the tools you need and most likely it is not only a tool to align reads. You might want to read reviews, tutorials and guidelines of the analysis on the dataset type you were given. Biostars is a good place to find answers to your questions. And there will be many.
The dataset I have is actually single end 5C reads! There are effectively not many tools to use for single end 5C data! So BLAST seems like the most viable option! Thank You
I think you'd rather try bwa or bowtie for the aligning, then can follow up with a pipeline like this https://genomebiology.biomedcentral.com/articles/10.1186/s13059-015-0806-y
I did try that! But I kept getting * in my SAM output! Is there any other way this could be done? I have been struggling with this for a while now. Thank You
Which tool did you try? How long are your reads?
I tried BOWTIE 2. My reads are all 80bp long. Any suggestions??
Hello bharath.mh24!
We believe that this post does not fit the main topic of this site.
It is unclear what is being asked here.
For this reason we have closed your question. This allows us to keep the site focused on the topics that the community can help with.
If you disagree please tell us why in a reply below, we'll be happy to talk about it.
Cheers!