Tutor wanted for remote sessions

[newkid]

Posted this in Bioinformatics too:

Hi everyone,

I was looking for a tutor that would be willing to sit down for a few one hour long sessions, once or twice a week. These sessions would be to explain and walk me through a few important concepts involved in genome assembly (de novo, and via reference). We would be working with publicly available data sets in a linux environment over gtalk/skype.

It would be ideal to have someone that has a considerable amount of industrial experience in bioinformatics with a broad understanding of computational biology.

I would be willing to pay a reasonable rate via paypal or any other agreed medium! I'm on Pacific standard time, and am available weekends or late evenings.

A few sample questions--if you can answer these off the top of your head you're in great shape:

1. What is a k-mer value?
2. How big and what are the file formats generated from a illumina hi-seq machine at 30x coverage? (for e. coli...)
3. How are data sets trimmed? (popular programs to do so)


Cheers!

[upendra_35]

Posted this in Bioinformatics too:

Hi everyone,

I was looking for a tutor that would be willing to sit down for a few one hour long sessions, once or twice a week. These sessions would be to explain and walk me through a few important concepts involved in genome assembly (de novo, and via reference). We would be working with publicly available data sets in a linux environment over gtalk/skype.

It would be ideal to have someone that has a considerable amount of industrial experience in bioinformatics with a broad understanding of computational biology.

I would be willing to pay a reasonable rate via paypal or any other agreed medium! I'm on Pacific standard time, and am available weekends or late evenings.

A few sample questions--if you can answer these off the top of your head you're in great shape:

1. What is a k-mer value?
2. How big and what are the file formats generated from a illumina hi-seq machine at 30x coverage? (for e. coli...)
3. How are data sets trimmed? (popular programs to do so)


Cheers!

Hi,

I am from CA as well and though i am not from Industry i have fair bit of RNAseq experience. I have done lots and lots of denovo and reference based assemblies and after undergoing through lots of pains and thrills i am very comfortable dealing with any kind of denovo and RB stuff.

Regarding your questions:
1. K-mer is the seed/word size of length k observed more than once in a sequence.

2. Assuming you are sequencing your library with SE of 50 bases you only need 3 million reads to get a coverage of 30X. So that is 1/50 of the lane in hi-seq.

3. I assume you are asking about the ways to trim you reads. If so there a many number of ways to do so and one popular tool is FASTX tool kit.

Thanks

[Jan_R]

Here are my 2 cents:

1.
... or in the dataset. When you can find exact same sequence very often in your raw data while you are checking its quality (e.g. with FastQC), it is most probably an artifact. Sequencing primers, adaptors. But it can also be rRNA. Blast such sequences and find out. If you do not like them, find ways to extract them.

2.
... Illumina files usualy come in the FASTQ format: http://en.wikipedia.org/wiki/FASTQ_format
The size of the files would some hundred MB

3.
... I also highly recommend the FASTX toolkit. Combining these tools is perfect to get your data in shape

[flyboyleo]

3. I thought 'Trim Galore' is better for integrated Cutadapt and FastQC to consistently apply quality and adapter trimming to FastQ files together

[Jan_R]

3. I thought 'Trim Galore' is better for integrated Cutadapt and FastQC to consistently apply quality and adapter trimming to FastQ files together

Good that you mention cutadapt.You can use both
I used cutadapt in combination with the FASTX tools. The FASTX Barcode Splitter seems to be not so versatile.

[newkid]

Position is still open

PM me, if it's of any interest.

[SES]

1. K-mer is the seed/word size of length k observed more than once in a sequence.

Just to be thorough, I'll add that a k-mer does not have to be observed more than once. Any k-mer that is observed exactly once is considered unique and the ratio of these to all other k-mers in a data set is a metric frequently used in comparative genomics.