Hi All, Thought it would be better to just ask the experts. So I, as a PhD, managed to convince my supervisory team that I needed to get some transcriptomics data. And, brilliantly, through saving money left, right and centre managed to get enough money to pay for the bioinformatics service with the sequencing. I had presumed this would come with normal quality assessments, assembly (to genome), differential expression and functional annotation/gene ontology to make sense of the samples.
However, I did not receive any functional annotation or gene ontology. Instead, I received data that said Gene1579836 was unregulated in sample x.
When I complained to the company, they said that because this was a poorly annotated genome, this was the best I was going to get. What I want to know is, is this normal? I mean I have my data, and I through much trial and error am blasting it myself and getting some nice results, but this has set me back about 6 months in work while waiting for the results that I can't really use since i'm still trying to get all the information to interpret the results.
Any guidance on this would be greatly appreciated. Given, that I am having to rerun the analysis again myself, I feel like i should get some of my money back.
Ask them what those genes are? If they have given you diff exp according to some known assembly, they they must tell you what are they. On the other hand, I'm afraid that they might have done some de-novo assembly of genome, created a gene-model and based on that, have given you the diff-exp. If your genome has not been assembled yet, indeed it's a tough job to do diffExp and then to interpret it functionally. Your another bet is to find a closely related organism with sequenced and annotated genome, and map your sequences there. You might get some interesting orthologs. up/down-regulated.
Which genome?
Didn't specify upfront (contract signed by both parties) what would be done and which wasn't possible?
Anyway, do you have access to the raw data (I sure hope so)? Because doing the analysis on your own might, in the end, be less painful.
Agree with @Wouter about the contract thing. See the finer details what they promised and if they are fulfilling it or not
I have asked and there answer was that I will have to do this myself (identify the genes that is) as I know the genome resources available better then them. The genome was the rainbow trout Oncorhynchus mykiss (no de novo assembly) and as for the contract, it was an extremely simple contract. The bioinformatics simple stated that they would do: (a) the data QC, basecalling and formatting; (b) alignment to reference and (c) differential gene expression. As a straight up biologist with minimum experience in this, I made the presumption that differential expression of genes would also include functional information otherwise i would not be able to give it context. I;m now worried that this is just an expensive error on my part due to my inexperience. And yes, luckily i insisted that i have access to the raw data otherwise i'd be extremely angry at this point. I'm currently blasting against the salmonid species and getting some nice results, but I just really wish i had been told this in advance rather then have to figure it out myself.
Please use
ADD COMMENTorADD REPLYto answer to previous reactions, as such this thread remains logically structured and easy to follow. I have now moved your post but as you can see it's not optimal. Adding an answer should only be used for providing a solution to the question asked.So, it looks like they did exactly as was specified upfront.
I'm not sure what you are blasting?
Many thanks for the reorganisation @Wouter. Yes, that is what I thought with regards to the service they provided. I made presumptions about what differential expression implied in the contract but didn't check to be certain that what I thought they were doing was actually what they were going to provide, hence my situation now. It is proving to be a costly experience that i just have to chalk up to my inexperience in this area.
In reply to your question, As I have the raw data, I have reanalysed the data using CLC workbench and am blasting the deferentially expressed data identified through this process.
I'm working on human genetics (and transcriptomics) so I'm spoiled by having a good annotation and reference genome. But (I might be wrong) this looks quite okay to me. There is also a GFF file available. Do you have someone in your institute who could guide you with this? With precise questions, you can always come here, but this looks like something that could you use more support with. It probably also depends on how much time you want and can invest in this. But for differential expression analysis there are excellent guidelines available, to start with I would like to point you to the Bioconductor RNA-seq workflow.
I am extremely lucky in that there is people i work with who are familiar with this type of data and are acting as a sounding board for when my knowledge fails me. But many thanks for the offer of assistance.
I am also lucky in that to truly understand the data, I have completed a few online courses that i thought could prepare me. However, these were based on well annotated and studied reference genomes.
My issue as always with my work is time. I had allocated 6 months to get this work done, but my contract is up in October and this is already on month 7 of the study. I have recently managed to obtain a small grant from the uni and have bought BLAST2GO to speed up the functional annotation. Many thanks for your help. I am now well warned about checking thoroughly what is actually provided in the service i am providing and not making assumptions.
Just for clarity: with BLAST2GO you are probably going to perform functional annotation. When you ask for differential expression you just test, for any gene model, if expression differs between conditions. There is absolutely no implication about functional annotation. However, I am surprised the company didn't offer to perform the functional annotation for a small additional fee.