Transcriptomics - Science topic

The kit was pretty user friendly but it didn't work well for my particular samples (leaves with high phenolic content).

NEB will often send a small sample-sized kit of many of their products. You could reach out to their customer service as well and ask if anyone has used it for your organism.

Depending on the type of your experiment, read duplication is to be expected and deduplication is not necessarily beneficial. You might want to have a look at this article: https://www.nature.com/articles/srep25533

Dear Stefanie,

from my experience there are scenarios where cells express genes that allow for the degradation of potential substrates which might not be present in the medium as part of a general stress or starvation response. This is the case for example in Bacillus subtilis which expresses genes of the DegU regulon comprising the upregulation of degradative enzymes like (proteases, peptidases, levansucrase, etc.) upon carbon depletion or entry into the stationary phase.

Also catabolite control or repression can mediate such a response (e.g. via cAMP in E. coli oder CcpA in B. subtilis).

Best

Michael

Hi Laura,

you have free tutorials in Galaxy web:

in which you can find the process to analyze RNAseq data. Then you can perform different analyses depending on your experiment design.

The average brain weight of a rat is 1.5-2 g, depending on the body weight and the age.

TRIZOL is a good option for RNA isolation; typical yields from brain tissue (mg RNA/mg tissue) are 1‑1.5 mg.

Transcriptomics and comparative genomics are the most standard techniques in NGS, and I would say "easy to interpret" comparing to any other technique such as CHIP-seq, ATAC-seq and many others.

First load the Robject file and see what it contains.

how to open an Robj file in r? - Stack Overflow

You can use ls() to list all the object in it.

For most of the analysis, there are libraries already written, so you just need to fine the type of analysis you want to conduct, look for library, install it and follow the documentation.

Certainly! I'd be interested in collaborating on this project. Please feel free to share more details about the research, and I'll be happy to discuss further. You can reach me at kdgb2007@yahoo.fr. Looking forward to potentially working together on reporting the results in a paper.

Hi Glauco,

Are you aware of the Xena browser (https://xenabrowser.net/)? There are at least gene expression, mutation and methylation data per patient for the TCGA cohorts. Unfortunately I am not sure about the TE data, since I am not working with that but I think that Xena would be your best bet.

Hi Pratanu,

have you checked if the parameters of featureCounts were set properly (correct annotation used, strand orientation, etc.)?

I would have a look at the bam files using a genome viewer, like IGV and check if there are reads at certain genes that I would expect to get reads for sure.

I think the gene ID in your GTF or GFF3 files you used for constructing the alignment index might not be the transcript ID including splice variant, which cause multiple alignment to 1 gene. I think you'd better use the genome annotation file and sequence file (gtf and fa) file from the ensembl (with gtf available at https://ftp.ensembl.org/pub/release-110/gtf/homo_sapiens/Homo_sapiens.GRCh38.110.gtf.gz and fa at https://ftp.ensembl.org/pub/release-110/fasta/homo_sapiens/dna/Homo_sapiens.GRCh38.dna.toplevel.fa.gz) or download pre-bulit index from the website or your alignment tools.

We welcome research articles, reviews, perspectives, commentaries, and clinical trials that discuss both basic and translational research as well as therapeutic perspectives in cancer from the view of metabolism.

Greetings! I'm delighted to share my insights and experiences with you in the realm of transcriptomics, immune responses, and machine learning. When comparing the performance of random forests and decision trees in predicting immune responses using transcriptome data, I've found the following performance metrics to be most informative:

When working with complex gene expression data, it's essential to consider the biological relevance of the performance metrics. For instance, in some cases, a high accuracy might not necessarily translate into biologically meaningful results. Instead, focus on identifying models that capture the underlying biology effectively, such as those that distinguish between different immune cell subtypes or predict functional pathways.

To optimize model parameters and prevent overfitting, follow these best practices:

By considering these performance metrics, biological relevance, and optimization strategies, you'll be well on your way to developing robust and reliable machine learning models for predicting immune responses using transcriptome data.

It is better to consider batch as a factor in the design formula. The tximport pipeline proposed by Michael Love himself offers the most useful solution. Please have a look.

I think a small server is a better choice for processing bioinformatics data analysis as it is cheaper and more convenient. This is because many analyses can take a long time, and MacBook do not have good heat dissipation.

Any transient transfection can be titrated to achieve a desired expression level. The CMV promoter is very strong and will work in most cell types. The mRNA and protein levels for your gene will vary depending on RNA stability, size, translational efficiency, etc. So, by transfecting varying amounts of your expression vector mixed with something inert, like Salmon sperm DNA, you can empirically determine how much you wish to express. For 100mm plate we use 6ug DNA with lipofectamine or fugene. But that can be 100 ng of vector with 5.9 ug inert DNA. Vector can be anywhere from 100ng to 6 ug.

Dear Kian,

have you tried improve your purity by FACS? It´s fairly easy to choose markers to distinguish MSC & macrophages and sort highly pure populations.

The shift between two opposite direction of change is the rule for regulatory genes that work as 'toggle.switches' in which the biphasic alternation of two conditions is the basis for a sort of digital control of biological regulation. It is not by chance that a great part of toggle-switches are retroviral origin sequences that mirror the lytic-lysogenic phases of viruses and phagi, see:

Oh, I see.

Thank you anyway!

I am not sure if could help, but have a look on https://www.omicsnet.ca/.

How does your institute financially support the applicants?

There are several reasons why frozen cells are not typically recommended for single-cell RNA sequencing:

To minimize these issues, researchers often prefer to use single-nuclei RNA sequencing for RNA profiling of frozen tissues, as this approach allows them to analyze the RNA from individual nuclei rather than whole cells. Single-nuclei RNA sequencing can be performed on frozen tissues, and it has the advantage of allowing researchers to profile the RNA from a large number of cells without the need for cell isolation or culturing.

If you need to freeze cells for RNA sequencing at a later time, it is important to handle the cells carefully to minimize the risk of RNA degradation and cell death. Some general tips for freezing cells for RNA sequencing include:

Overall, it is generally recommended to use fresh cells or single-nuclei RNA sequencing for RNA profiling, rather than frozen cells.

Hi Eugene,

I'm currently in the process of optimizing and planning a pilot experiment using the DNTR method explained below:

"A Highly Scalable Method for Joint Whole-Genome Sequencing and Gene-Expression Profiling of Single Cells": https://www.sciencedirect.com/science/article/pii/S1097276520306559

Let me know if you like to brainstorm together.

Tu add up to my college Daniel Toro-Domínguez ,

Both approaches are good but they are different types of enrichment analysis.

in 1) you will be doing a GSA where the results might vary according to the arbitrary cutoff to define, in your case, the "targets with high predicted repression". Using a specific background of target genes was proposed by several authors to correct probable bias effects.

in 2) You will be doing a GSEA "threshold free approach".

or 3) Using GeneCodis providing it directly your list of miRNAs that can be:

- Diretly tested the miRNAs with specific annotations or gene to mirnas transformed databases using the hypergeometric test.

- Or if you want to analyse the targets, use the Wallenius test to address a gene selection bias known in GSA of CpGs, miRNAs and TFs.

I would perform all of them and chose the results that allows you a better discussion / conclusion.

I ended up writing a python script which produced a GMT file that was valid for a GSEA. Here is the code if anyone else needs to do this:

You can conduct RNA-Seq analysis (only in host or dual in both pathogen and host) to get the DEGs between infected and non-infected conditions and get a bit of idea regarding the genes and metabolic pathways which might be involved in inciting the disease response

I think there are 2 major approaches: 1. Marker genes - the idea is that a gene with high average fold change and appropriate adjusted p value between all clusters is uniquely representing a cell type. Markers can be “canonical” - surface proteins detectable with flow cytometry, or “literature based” - genes known to distinguish cells by type and validated in literature. This is a developing area with new publications new frameworks to characterize cells and assign them with a defined type ( )

The second approach is classification based. 2. Using large collections of cells (I.e. https://www.humancellatlas.org or even cell line experiments with specific ceo type gene expression) we train a model to predict class assignment for new gene expression data. Such trained models rely on various numbers of datasets and features (hundreds) and more complex patterns than just logFC and adj. p-value. These are methods like celldex (https://bioconductor.org/packages/release/data/experiment/html/celldex.html).

Both methods have limitations in practice - many clusters can be assigned to several types of cells based on “good” logFC and p-value, so the user might choose the top values or go for unique cell type not present in other clusters. Since methods like Seurat can assign a cell type to each cell, you can also calculate proportion of cell types in a cluster and use that.

Automated annotation can also fail to assign a good cell type to a given set of cells. since there are many known types of cells and new variations are often found, combining these approaches and performing additional manual examination of marker genes, literature and expression patterns is typically required.

Hope that helps! We’re adding tutorials on this topic on our OmicsLogic portal: https://learn.OmicsLogic.com

Markus Glaß thank you!

Simply download the raw counts as .csv and process it using DESeq2 or limma-voom in R. For limma you will have to normalize your counts first. DESeq2 is easier. You will get log2FC values along with adj. p value. If you just want to check their expression for wet lab experiments. You can also calculate the fold change without checking for statistical significance. Just average the counts of control samples and experiment samples and FC= exp/control .

i am working on cardiomyopathy patients Blood samples . and wanted to do miRNA sequencing can some one please suggest how many millions reads i need to sequence 20 millions or 30 millions and also please suggest the platform as well .

Agniva Saha the problem was RNA integrity. I don't have my full protocol notes but briefly, cell lysis was performed in a 'gentle' buffer consisting of hypo-osmotic sucrose and 0.5% Triton X-100, agitated at a low temperature. I supplemented with superasin. Following lysis (which was incomplete after 30 minutes) the buffer was supplemented with 1.5% SDS and sodium chloride, Proteinase K was added, and Prot K digestion performed at room temperature for 45 minutes.

This whole protocol is a huge tradeoff between the needs of RNA (low temperature, rapid separation from other cellular components) and deproteination using Prot K (ideal reaction temperature 50-65C, and for deproteination of DNA after ChIP we do 4-hour reactions to get rid of all the protein.) The result was that I isolated less than 1/3 of the typical yield of low-integrity RNA (RIN ~5.5-6.5).

Compare this to phenol, which is such a strong protein and nucleic acid denaturant that it results in a) near complete lysis and b) near complete protection of RNA from endogenous RNases within seconds of efficient sample homogenisation. I don't really think there's an alternative that goes close.

I will say though, every method I mentioned above for avoiding phase separation results in no loss in RNA integrity if performed properly. I haven't done a head to head yield comparison, but my routine RNA extraction protocol involves Nucleozol supernatant mixed 1:1 with ethanol and run through a column, and the yields are better in my hands than alcohol precipitation of RNA, with better purity.

Not only it is much easier to work with transcriptomics but transcriptomics allow you to take into consideration the role played by non-coding sequences that are the majority of the genome and very often play a crucial role in biological regulation see for example:

This article might help you to MP analysis. You can get a detailed idea from the corresponding author of this paper.

Jabed Hasan, can you please help him to characterize the MP.

Hi Anna,

There are several freezing protocols in internet for mammalian cells.

My general protocol for freezing is like that, it should work with other mammalian cells:

*Make Cryoprotectant solution with following final composition:

10% DMSO (D2650, Sigma-Aldrich),

50% Fetal Bovine Serum,

40%-Cell culture medium without FBS,

1% antibiotic/antimycotic mixture (Penicillin/Streptomycin mixture).

*Suspend cells in cryoprotectant solution at a final concentration (5 million to 10 million cells/ml)

*Place the cell suspension in cryotubes (fill 70% only)(The cryopure tubes (cat# 72.377.002, Sarstedt or from any company)

*Place The cryopure tubes with cell suspension in Mr. Frosty™ freezing container (Cat# 5100-0001, Thermo Scientific)

*Place Mr. Frosty (with cryotubes) at -70˚C to -86C for 24h

*Either Transfer Cryotubes to liquid nitrogen, you can store for upto many years.

or

transfer these cryotubes tubes in cold boxes (-70 to -86C) and leave them there (-70 to -86C) for storage upto 1 year only

Thawing of cells:

Take cryotubes out of LN2 or Freezer (-70 to -86C) and thaw them at 37C water bath, for 5 to 10 minutes till the suspension melts.

Transfer the cell suspension in cell culture medium (you decide) in cell culture flask (you decide) under sterile hood.

Grow cells for a passage to recover from freezing shock than use for your RNA seq or other experiment after first or second passage

(I assume you know cell culture method, otherwise you can take help from anyone doing cell culture)

Good luck, any question, you can ask

Subhash

Check out Aymoz et al. (2018) "Timing of gene expression in a cell‐fate decision system" (Mol. Syst. Biol. 14, 4, e8024) for a nice study on the timing of transcriptional responses mediated via MAPK to yeast pheromones, a system similar to antigen-induced gene activation pathways.

The cascade leading to the transcriptional factor activation is short, in the case of the canonical ones (NFkB, IMD). The accumulation of the transcripts is another subject matter. The structure of the promoters of AMPs suggest non immune related transcription factors are likely to bind and interfere with the specific transcription. HIF binding site, for instance, is frequently found nearby these genes, suggesting a potential hindrance upon redox stress.

Correlation AnalyzeR: functional predictions from gene co-expression correlations

BMC Bioinformatics volume 22, Article number: 206 (2021) Cite this article

Abstract Co-expression correlations provide the ability to predict gene functionality within specific biological contexts, such as different tissue and disease conditions. However, current gene co-expression databases generally do not consider biological context. In addition, these tools often implement a limited range of unsophisticated analysis approaches, diminishing their utility for exploring gene functionality and gene relationships. Furthermore, they typically do not provide the summary visualizations necessary to communicate these results, posing a significant barrier to their utilization by biologists without computational skills.

Background Almost two decades after the completion of the Human Genome Project, the functionality of many genes remains largely enigmatic [1]. Many such “enigmatic genes” have immense biological significance, exemplified by the associations of thousands with cancer outcome [2]. Even genes which are well-characterized often play unexpected roles in different biological contexts (e.g., EZH2 is both a tumor-suppressor and an oncogene in different cancers [3]). Gene co-expression correlations provide a robust methodology for predicting gene function, as genes which share a biological process are often co-regulated [4,5,6]. Similar insights can be gained from using protein interaction (for example STRING [7] and InterologFinder [8]), phenome data, or even the combination of both [9]. Irrespective, generating expression data remains a cost-effective approach and co-expression analysis remains a prominent tool for exploratory systemic evaluation, largely because it is capable of considering gene co-expression across the genome. However, the applications which have been developed for such inference are hampered by key limitations. Tools like COXPRESdb [10] and GeneFriends [11] calculate gene set over-representation on an arbitrary number of co-expressed genes. Alternatively, GeneMANIA [12] and GIANT [13] construct co-expression networks and calculate gene set over-representation on an arbitrary number of nodes. Neither approach is sensitive to the genome-wide distribution of co-expression correlations or, with the exception of GIANT, differences between tissue/disease conditions. Furthermore, these functional predictions are limited in scope and do not generate relevant, user-friendly visualizations, limiting their utility for biologists without bioinformatics skills.Recently, Lachmann et al. introduced ARCHS4, a database with thousands of standardized RNA-Seq datasets [14]. We re-processed these data, calculating co-expression correlations with respect to tissue and disease (cancer/normal) condition and provided the results in a publicly accessible database. We now present Correlation AnalyzeR, a user-friendly interface to this co-expression database with a suite of tools for de-novo prediction of gene function, gene–gene relationships, and biologically relevant gene subgroups to facilitate discovery of novel relationships within genes of interest.

Hi

RNA later is generally used when the whole the integrity of the RNA inside the tissue needs to be maintained, if it is not possible to freeze in the Liquid Nitrogen, immediately. It is possible that RNA later may also cause problem in RNA isolation, if not removed properly. Its removal is relatively easy from the surface of whole tissues, than if it is added to the homogenized tissue. For your experimental needs, the TRIZOL/TRI reagent may be a better good option. It maintains the quality of RNA in the homogenized tissue. And it should be useful of high-throughput isolation of RNA

Pragya Tiwari Mam i am also going to use the same kit to do 3` RACE.I am planning to do pcr using 3` outer primer and 5` gene specific primer. Can i amplify my 3 kb size gene of interest using this 3` RACE only.Kindly reply

Yes indeed... Polyadenylation controls many things, but mainly the stability of the transcript. It is well known in mammals that for instance those proteins involved in signal transduction have the tendency to harbor weaker polyadenylation signals, which conditions the stability of the mRNA and the protein levels.

Hi,

I recommend using NanoDrops and agarose gel electrophoresis to check the quality and integrity of your extracted RNA sample, so you can use it if it has good quality and integrity.

The main consideration when choosing a lysis buffer is whether the chosen antibody will recognize denatured samples. When this is not the case, it will be noted on the antibody datasheet, and buffers without detergent or with relatively mild non-ionic detergents (NP-40, Triton X-100) should be used.

Hi Kleran

I think you followed the support (https://support.10xgenomics.com/spatial-gene-expression/software/pipelines/latest/advanced/references) and I see 2 points that could be at the origin of your problem:

- the first one is the genome you selected, is it mm10 genome?

- the second one is that data you downloaded must be compatible with STAR aligner, a point you need to dig in...

all the best

fred

You may follow this pipeline.

Let me know if you are interested to outsource to my lab www.eminentbio.com

Differential expression of Transcriptome means the difference between the expression of genes whether up-regulated or down-regulated. The read counts are deduced by mapping followed by gene annotation. Read counts means the number of time a sequence overlap to the genomic feature such as gene or transcripts. These all read counts of given samples (eg. control and treatment)are provided to the respective tool in form of a matrix which is then compared by log base 2 (control /treatment). P-value is the probability of occurrence of the test randomly. A P-value less than 0.05 is considered best.

Honestly, this tech is still in its infancy, and is also _suuuuuper_ expensive. Most people are hard-pressed to make use of the gigabytes of sequence data they get from a single experiment, let alone take that forward to N=3 or more.

That isn't to say what you're proposing is impossible, or even impractical, but more to suggest that for such specific questions there are usually cheaper, more specific solutions. You absolutely can compare multiple RNAseq datasets, and adding a spatial component to this should also be possible provided your segmentation/tissue designation is solid enough (or you have cell-specific markers to aid comparisons), but really...I think you need to carefully formulate your question and work out exactly what this very, very expensive (but very, very neat) approach can do that other more conventional methods cannot.

Also, obviously I am jealous of your budget and resources, but that kinda goes without saying. ;-)

Hi,

It depends on the size of the datasets in your study. e.g. for metagenomic data analysis, you can analyze your data with a limited memory(like qiime2,metaquast,dada2,etc), but for transcriptomic data analysis, most of the program (STAR, bowtie, bwa, etc) require a substantial amount of memory for mapping, indexing of the genome (specially Human, Mouse). For example, STAR aligner requires minimum 150GB of RAM for indexing human genome. Currently we are using Dell Power Edge (R940XA) rack server for our transcriptomic, metagenomic study. Briefly, the server has Intel Xeon Gold 5220 processor with 72 cores (number of core per process:18, 4RU,), 1 TB RAM (DDR4 SDRAM with ECC), 24.75 cache memory (L1+L2+L3), 10TB storage with RAID configuration.

Regards;

Anupam

Adding to the above points, this is a common practice and often adopted by many researchers, however, the problem only arises when publishing the research without providing any data to the public repositories.

Hi Edgardo,

Technically, you still can perform GSEA even though you are using non-model plant. Some of the necessary steps/files are as follows:

1. a gmt file. I don't reckon that there is a readily available gmt file for your plant. So, you may be able to use the gmt file of the closest plant relative to your species. This publication (https://pubmed.ncbi.nlm.nih.gov/23632162/) and this database (http://systemsbiology.cau.edu.cn/PlantGSEA/download.php) may help.

2. You will need to map your gene ID to match that of the gmt file. To do so, you will need to annotate the transcriptome which you have assembled. I am not familiar with the annotation process, but this manuscript may help ( ). After the transcriptome annotation, you can map your transcripts to genes. If the gene ID matches that of the gmt file, the files can be used directly for GSEA. If not, you will need to remap the gene ID before GSEA. BiomaRt (https://plants.ensembl.org/biomart/martview/76f17c602cf7f4cb9adc75e9a97c87df) may be helpful in this case.

I hope this helps.

Regards

Hong Sheng

Anything which makes sense and provide/help in visualizing the thing you write and describe in paper.

I guess if you are the author, it should be you to decide what to put in your paper. If you are not sure, ask co-authors, if you have any.

Matthew Edward Thornton

Is there any method to work with NONCODE in R?

Hi

I suggest using if and else in lapply.

for example:

if(nrow(coulmn1)> 30) {

x <- with(data, cor(a, b))

}

else {

x <- 0

}

This is a good solution when the number of samples are small.

I appreciate Sevcan Atay for the valuable topic. Would be interested to know as well.

Hi Mabrur, it is a very well-known fact that transcriptome expression is highly variable and is age-specific, time-specific, gender-specific, organ / tissue specific, environment-specific and also differs individual to individual and the same matters for plants too. It all depends what you are trying to find in the transcriptome and does that entity have any impact on its expression profile due to any or all of the above-mentioned factors and based on this decide whether it is better to go for transcriptome mapping or transcriptome assembly in your case. Also you need to take in account several other factors like adopting proper extraction protocols, yielding good proportion of the transcriptome you are interested (coding / non-coding / whole), good library preps, and using proper transcriptomic controls for judging asay variations, use substantial amount of replicates (biological and technical both) etc. Only then you can rely on the results

of the transcriptome experiments.

Regards...

Kindly check http://www.bio.vu.nl/thb/course/ecol/MetzMyli2008.pdf

Hi Albert

I think this article:

"Extracellular Microbial Metabolomics: The State of the Art"

Good Luck

The ideal is you produce your own control (ideally, at least three replicates to each condition + three replicates of control samples), even if you find fastq files of tissue transcriptomes.

The library building process and the sequencing itself can introduce some "biases". So, if you have all sequencing samples produced at the same time with the same conditions, they will be more comparable and your results more reliable.

As Amir Vajdi said if you have no disease or stress or other condition and you want to compare the expression between tissues, you can compare one tissue to the other.

Dear Frankie;

I usually use a trick on MetaboAnalyst web server (www.metaboanalyst.ca). I dosent have direct conversion system. therefore, put the names in the enrichement section. The database will offer you most of the HMDB IDs.

Browse to the homepage

There is a red line " Welcome>> click here to start << "

Then use the "enrichment analysis" section.

Best

Hello Claudia Zagami , Peter Kilfeather ,

Our CRO lab (Arrayjet Ltd.) have successfully printed oligos on Codelink slides for use in spatial transcriptomics projects, mentioned in this Nature Comms paper: https://www.nature.com/articles/ncomms13182

The slides do have a limited shelf-life, so your problem may lie there.

If either of you are still working on this, I'd be pleased to arrange a call/webchat with our Application Scientist and Project Manager who extensively optimised the printing protocol.

I can be reached directly at: shawkings@arrayjet.co.uk or simply drop me a line here and I'll do what I can to help.

Best wishes,

Sam

Frederic Lepretre Probably. By the way, I found the corresponding author's email by the google search and got the information that I need.

We have used multiple tools to analyze TCGA data. Links have been attached in the paper for easy reproduction. It will help for a good start with TCGA

Hi colleague

Find the following URL, may help you:

Regards..

northern blot analysis

Informatics analysis would be the limiting factor. If you are going with commercial sequencing, make sure it is included.

I will also recommend following the recently published article in Nature journal.

You can use a simple script which I have written in python here for GO enrichment with a list of input genes:

(just run it in google colab)

Hi again, Ali Motahharynia

I think you should plot your points using error bars with replicates. Unfortunately, you can not conclude anything because you don't have replicates (at least three per time point).

Now, assuming that your points are the average of some replicates, what is exactly your question? I mean, they look similar in their evolution, but I don't know what you want to know about them. Could you be more specific in your question?

And finally, an advice for the future, please put names in your plots.

Regards,

Hi,

I Think 100% methanol fixation would be useful for getting rid of infection causing virus. Apart from that you could try fixing tissue by using 4% PFA or 10% Neutral Buffered formalin. That would be effective! Also you may try working with RNA later. It should destroy viral protease

Thanks a lot for your response Karol and Ireneusz!