Inexperienced Question

cambridge101 · 12-18-2014, 12:50 PM

Hello,

I have a project that requires access to an "online cancer related DNA database resource."

Unfortunately, it doesn't look like I can get access to CGHub. Would someone mind helping me? (i.e. perhaps suggest one that I can access)

Thank you!

AntonioRFranco · 12-18-2014, 01:07 PM

www.cartagenia.com
You don't mention if it should be of free access or not

cambridge101 · 12-18-2014, 02:06 PM

AntonioRFranco - Thanks for the reply. Yes, free.

GenoMax · 12-18-2014, 02:43 PM

http://www.cbioportal.org/public-portal/

There are at least a couple of companies that offer different views of the TCGA data via web. Search on SeqAnswers.

GenoMax · 12-18-2014, 02:49 PM

http://seqanswers.com/forums/showthread.php?t=48485
http://seqanswers.com/forums/showthread.php?t=40610

cambridge101 · 12-18-2014, 02:50 PM

GenoMax - Thanks for the reply. I probably should have mentioned the before: select ten (10) DNA sequence strings of length at least 1Mb related to a cancer gene from ten different individuals. Make sure the sequence data is in FASTQ format and stored in one file “DNA.fas”.

GenoMax · 12-18-2014, 02:54 PM

Now you say

You are not going to find DNA sequences in fastq format that are 1Mb in length unless you assemble them yourself preserving the Q-scores (assembly programs do not include Q-scores in final sequence).

cambridge101 · 12-18-2014, 02:57 PM

I have a few days to work on this project. I'm lost.

GenoMax · 12-18-2014, 03:01 PM

Tell us more about what the entire project is about. DNA sequence is just a part of it?

cambridge101 · 12-18-2014, 03:09 PM

GenoMax - Thanks again for your suggestions and assistance. I'm just a little reluctant to say more about this project right now. I hope you understand.

cambridge101 · 12-18-2014, 03:22 PM

Though I don't feel I was doing anything unethical, others may disagree. Therefore, I deleted the complete requirements to avoid any conflict. Thank you.

GenoMax · 12-18-2014, 03:42 PM

I think you should delete the description from the post above. This appears to be a class project/home work?

Perhaps you should ask whoever assigned the project if they are certain about the fastq requirement.

cambridge101 · 12-19-2014, 05:06 AM

Thanks for your suggestion. I don't think I'll delete it[Did decide to delete]. I don't feel that I'm doing anything unethical. I hope that it's clear that I'm not asking for anyone to complete the project for me. I'm just having a hard time finding that FASTQ data I need.

GenoMax · 12-19-2014, 05:16 AM

One place to look for long sequences (in fastq format) will be here: http://www.pacificbiosciences.com/ne.../publications/ I see only a couple of cancer related publication and they are from 2012 (when the reads were not as long as they are today).

If you drop the cancer requirement then you will get some really long reads here: http://blog.pacificbiosciences.com/2...d-shotgun.html They are not going to be 1Mb so you would still need to do some assembly.

cambridge101 · 12-19-2014, 05:34 AM

Quote:

Originally Posted by GenoMax

One place to look for long sequences (in fastq format) will be here: http://www.pacificbiosciences.com/ne.../publications/ I see only a couple of cancer related publication and they are from 2012 (when the reads were not as long as they are today).

If you drop the cancer requirement then you will get some really long reads here: http://blog.pacificbiosciences.com/2...d-shotgun.html They are not going to be 1Mb so you would still need to do some assembly.

Thanks! I'm not sure I can drop the cancer requirement. I checked: http://www.pacificbiosciences.com/ne.../publications/ I couldn't locate the data I need.

I'm open to other suggestions if anyone has any.

SNPsaurus · 12-19-2014, 08:41 PM

I only vaguely remember the details of the project from your deleted post, but if I were to guess what a reasonable assignment would be, it would be to select a cancer gene, then extract 1 Mb of sequence around the gene from ten different individual genomes, then analyze those 1 Mb regions for the various things asked for in the post.

You aren't going to find 1 Mb fastq reads, but you can find different individual genomes, or even different "cancer" genomes. You can definitely identify genes related to cancer. As others have said, I'd check back with the assigner of this project for clarification.

edit: I teach an upper level course in genomic methods and analysis, so am definitely curious what this assignment is about!

GenoMax · 12-20-2014, 05:42 AM

I like SNPsaurus' interpretation. 1 Mb (total amount of data or length of region covered) worth of fastq reads in and/or around a cancer gene makes sense.

@SNPsaurus: Main input for the assignment is still in post #6. The rest of the assignment was informatics goals.

This will entail a significant amount of work (data collection part) and I hope the assignment has an appropriate amount of credit (unless it is a PhD qualifier exam).

cambridge101 · 12-20-2014, 12:05 PM

Alright... Let's say my oncogene of interest is in the region of 11:15000000..16000000. Therefore, all I need is that region from 10 different people.

Problem:
Where do I find that data???

Any assistance is appreciated.

GenoMax · 12-20-2014, 01:59 PM

Look for studies that have > 10 samples (since you need 10) or take 10 samples from different cancer types.

http://sra.dnanexus.com/?result_type...q=tumor+exome+

http://sra.dnanexus.com/?result_type...q=cancer+exome

Richard Finney · 12-20-2014, 06:40 PM

The comments here have links to sequences for PUBLIC human cancers ...

http://www.homolog.us/blogs/blog/201...able-from-bgi/

BGI liver cancer
Seoul Genomic Medicine Institute lung cancer
Changhai Hospital prostate cancer
MD Andersen Asian Gastric cancer

I think the data is in NCBI's SRA

You'll need a lot of disk space and, if you're relatively new, a lot of patience.

Sadly, a "bam slicer" that cuts out the reads for a region isn't available; though they say some folks are working on it.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Heatmap question	lucer105	Bioinformatics	0	12-27-2013 10:01 PM
First Question	njt638	General	1	08-24-2011 12:56 PM
Question	cjose	Illumina/Solexa	4	08-11-2011 06:31 AM
question?	semna	Bioinformatics	7	12-20-2010 05:30 AM
question?	semna	Bioinformatics	1	12-17-2010 02:54 AM

12-18-2014, 12:50 PM	#1
cambridge101 Member Location: Boston Join Date: Dec 2014 Posts: 10	Inexperienced Question Hello, I have a project that requires access to an "online cancer related DNA database resource." Unfortunately, it doesn't look like I can get access to CGHub. Would someone mind helping me? (i.e. perhaps suggest one that I can access) Thank you!

12-18-2014, 03:22 PM	#11
cambridge101 Member Location: Boston Join Date: Dec 2014 Posts: 10	Though I don't feel I was doing anything unethical, others may disagree. Therefore, I deleted the complete requirements to avoid any conflict. Thank you. Last edited by cambridge101; 12-19-2014 at 08:08 AM. Reason: Information not necessary.

12-19-2014, 05:06 AM	#13
cambridge101 Member Location: Boston Join Date: Dec 2014 Posts: 10	Thanks for your suggestion. I don't think I'll delete it[Did decide to delete]. I don't feel that I'm doing anything unethical. I hope that it's clear that I'm not asking for anyone to complete the project for me. I'm just having a hard time finding that FASTQ data I need. Last edited by cambridge101; 12-19-2014 at 08:10 AM. Reason: Inconsistant with edit to earlier post.

12-19-2014, 08:41 PM	#16
SNPsaurus Registered Vendor Location: Eugene, OR Join Date: May 2013 Posts: 521	I only vaguely remember the details of the project from your deleted post, but if I were to guess what a reasonable assignment would be, it would be to select a cancer gene, then extract 1 Mb of sequence around the gene from ten different individual genomes, then analyze those 1 Mb regions for the various things asked for in the post. You aren't going to find 1 Mb fastq reads, but you can find different individual genomes, or even different "cancer" genomes. You can definitely identify genes related to cancer. As others have said, I'd check back with the assigner of this project for clarification. edit: I teach an upper level course in genomic methods and analysis, so am definitely curious what this assignment is about! __________________ Providing nextRAD genotyping and PacBio sequencing services. http://snpsaurus.com

12-18-2014, 01:07 PM	#2
AntonioRFranco Member Location: Cordoba, Spain Join Date: Feb 2013 Posts: 21	www.cartagenia.com You don't mention if it should be of free access or not

12-18-2014, 02:06 PM	#3
cambridge101 Member Location: Boston Join Date: Dec 2014 Posts: 10	AntonioRFranco - Thanks for the reply. Yes, free.

12-18-2014, 02:43 PM	#4
GenoMax Senior Member Location: East Coast USA Join Date: Feb 2008 Posts: 7,083	http://www.cbioportal.org/public-portal/ There are at least a couple of companies that offer different views of the TCGA data via web. Search on SeqAnswers.

12-18-2014, 02:49 PM	#5
GenoMax Senior Member Location: East Coast USA Join Date: Feb 2008 Posts: 7,083	http://seqanswers.com/forums/showthread.php?t=48485 http://seqanswers.com/forums/showthread.php?t=40610

12-18-2014, 02:50 PM	#6
cambridge101 Member Location: Boston Join Date: Dec 2014 Posts: 10	GenoMax - Thanks for the reply. I probably should have mentioned the before: select ten (10) DNA sequence strings of length at least 1Mb related to a cancer gene from ten different individuals. Make sure the sequence data is in FASTQ format and stored in one file “DNA.fas”.

12-18-2014, 02:54 PM	#7
GenoMax Senior Member Location: East Coast USA Join Date: Feb 2008 Posts: 7,083	Now you say You are not going to find DNA sequences in fastq format that are 1Mb in length unless you assemble them yourself preserving the Q-scores (assembly programs do not include Q-scores in final sequence).

12-18-2014, 02:57 PM	#8
cambridge101 Member Location: Boston Join Date: Dec 2014 Posts: 10	I have a few days to work on this project. I'm lost.

12-18-2014, 03:01 PM	#9
GenoMax Senior Member Location: East Coast USA Join Date: Feb 2008 Posts: 7,083	Tell us more about what the entire project is about. DNA sequence is just a part of it?

12-18-2014, 03:09 PM	#10
cambridge101 Member Location: Boston Join Date: Dec 2014 Posts: 10	GenoMax - Thanks again for your suggestions and assistance. I'm just a little reluctant to say more about this project right now. I hope you understand.

12-18-2014, 03:42 PM	#12
GenoMax Senior Member Location: East Coast USA Join Date: Feb 2008 Posts: 7,083	I think you should delete the description from the post above. This appears to be a class project/home work? Perhaps you should ask whoever assigned the project if they are certain about the fastq requirement.

12-19-2014, 05:16 AM	#14
GenoMax Senior Member Location: East Coast USA Join Date: Feb 2008 Posts: 7,083	One place to look for long sequences (in fastq format) will be here: http://www.pacificbiosciences.com/ne.../publications/ I see only a couple of cancer related publication and they are from 2012 (when the reads were not as long as they are today). If you drop the cancer requirement then you will get some really long reads here: http://blog.pacificbiosciences.com/2...d-shotgun.html They are not going to be 1Mb so you would still need to do some assembly.

12-20-2014, 05:42 AM	#17
GenoMax Senior Member Location: East Coast USA Join Date: Feb 2008 Posts: 7,083	I like SNPsaurus' interpretation. 1 Mb (total amount of data or length of region covered) worth of fastq reads in and/or around a cancer gene makes sense. @SNPsaurus: Main input for the assignment is still in post #6. The rest of the assignment was informatics goals. This will entail a significant amount of work (data collection part) and I hope the assignment has an appropriate amount of credit (unless it is a PhD qualifier exam).

12-20-2014, 12:05 PM	#18
cambridge101 Member Location: Boston Join Date: Dec 2014 Posts: 10	Alright... Let's say my oncogene of interest is in the region of 11:15000000..16000000. Therefore, all I need is that region from 10 different people. Problem: Where do I find that data??? Any assistance is appreciated.

12-20-2014, 01:59 PM	#19
GenoMax Senior Member Location: East Coast USA Join Date: Feb 2008 Posts: 7,083	Look for studies that have > 10 samples (since you need 10) or take 10 samples from different cancer types. http://sra.dnanexus.com/?result_type...q=tumor+exome+ http://sra.dnanexus.com/?result_type...q=cancer+exome Last edited by GenoMax; 12-20-2014 at 02:07 PM.

12-20-2014, 06:40 PM	#20
Richard Finney Senior Member Location: bethesda Join Date: Feb 2009 Posts: 700	The comments here have links to sequences for PUBLIC human cancers ... http://www.homolog.us/blogs/blog/201...able-from-bgi/ BGI liver cancer Seoul Genomic Medicine Institute lung cancer Changhai Hospital prostate cancer MD Andersen Asian Gastric cancer I think the data is in NCBI's SRA You'll need a lot of disk space and, if you're relatively new, a lot of patience. Sadly, a "bam slicer" that cuts out the reads for a region isn't available; though they say some folks are working on it.