Featured Resource: AxoBase

Doing great science depends on teamwork, whether this is within the lab or in collaboration with other labs. However, sometimes the resources that support our work can be overlooked. Our ‘Featured resource’ series aims to shine a light on these unsung heroes of the science world. In this interview, we spoke to the team behind AxoBase, a new platform providing a one-stop web resource for the axolotl research community.

Could each of you briefly introduce yourself?

Prayag: My name is Prayag Murawala. I am an assistant professor at MDI Biological Laboratory. I started my lab three years ago. Before that, I did my postdoc with Elly Tanaka at IMP Vienna. I’ve been working with axolotl for the last 13 years.

Jessica: I’m Jessica Whited. I’m at the Department of stem cell and regenerative biology at Harvard, in Cambridge, Massachusetts. I’ve been working on axolotl models for 17 years.

James: My name is James Godwin, and I’m an assistant professor at MDIBL. I’ve been working with axolotl since 2006.

Joel: My name is Joel Graber. I am the Director of Computational Biology and senior staff scientist at MDI Biological Laboratory. I’m a computational biologist. I’m originally a physicist and computer scientist by training, but I’ve been doing biology since 1996. My job is to manage the computational end of AxoBase, bringing axolotl-related resources together and making them available.

Why and how was AxoBase set up? 

Prayag: The Axolotl genome was assembled in two different laboratories several years ago — the Elly Tanaka laboratory at IMP Vienna, and Randal Voss and Jeramiah Smith at the University of Kentucky. Around the same time, there was a salamander meeting initiated by the community. From that meeting, a white paper was published, co-authored by James Monaghan (Department of Biology, Northeastern University), Crystal D. Rogers (School of Veterinary Medicine, University of California-Davis), Jeramiah Smith, Randal Voss, and Jessica Whited.

One of the things listed in the white paper is that we want to have a common platform. This has been discussed in the community for a while, but I think the lack of initiative was the major hurdle. At that time, and even today, there are many different websites, including the Axolotl-omics website, which I beta-tested when I was still in the Tanaka lab. After I took up the position at MDIBL, I was in contact with James Godwin and Joel Graber because they are also faculty here. James was trying to combine the list of antibodies that work with axolotl. I told James and Joel that I would really like to build a resource for axolotl research. We reached out to Jessica as well because of her experience with the axolotl transcriptome assembly. That was how this entire team was formed.  

Jessica: I previously trained in flies, so I was very used to having all the genomic resources, mutant and reporter lines easily accessible for researchers. I’ve seen a massive improvement over what existed when I first started in axolotl research, however, there’s still a lot of room for improvement. One of the huge issues is the usability of the existing data. When Prayag, James and Joel reached out about joining forces, I thought that was a really great idea, because it’s really important that the field coalesces to create these kinds of resources. It’s great that we have a cross institution initiative to make this happen. I’m happy to be part of it.

James: I have a similar story. I’ve also worked with mice and it’s just ridiculously easy to do anything genomic as the resources are very vast. But in axolotl, we’re still so far behind and the resources are fractured. I’ve been working with axolotl since 2006, when there weren’t a lot of bioinformatics tools available. You would email someone in the stock center in Kentucky and ask them to look up if they’ve got any genomic data that they can share on your gene of interest. The axolotl genome was published a few years ago, but for an end user, things haven’t moved much in that time. I want the axolotl model to grow, but without those genomic resources in place, we’re not going to attract people into the field.

What is AxoBase? 

Prayag: At the moment, we only have links to the Genome Browser and all the resources that are hosted on different websites. We also have an antibody list that James had compiled. The third page lists all the transgenic lines that are published in the field. And then we have a list of labs that are working with axolotl and salamanders in general. The last page is about how we want to develop this, and what are the different areas we want to reach out to. We intend to form committees with the help of the whole community to move it forward.

Joel: AxoBase is still very preliminary. Right now, it’s a community supported information portal for axolotl research. We’re making sure the labs know of each other, that they know of upcoming meetings and things of that nature. It is not yet a genome portal nor the knowledge base that we want it to be.

How can the community help and contribute to AxoBase?

Joel: We’re putting together a proposal, and we need members from the community to help build an anatomical ontology, maintain, and update the gene nomenclature. All of this has to be done in congruence with the external world.

Prayag: We want to integrate all the existing information in the field, including antibodies, probes for HCR in situ hybridization and guide RNAs. But ultimately, no matter how much computation power you use, the annotations are never perfect. The ultimate good quality of annotation comes from the users — the users have to verify that a gene really exists. In axolotl, there are a lot of duplicated pseudogenes. This will require a huge community effort. We require everybody’s participation, whether somebody developed a transgenic animal, used an antibody that is working, assembled the genome or transcriptome, or developed a time course analysis. Eventually, we want to plug this all into AxoBase, so that information we have generated is not lost and can be found easily, not only by existing axolotl researchers, but also for the next generation of scientists wanting to get into the field.

Jessica: One problem, at least in our lab, is the usability of the data from published big datasets. If you don’t have an in-house computational biologist who can help, then it’s really hard. I think we should also be aiming to have a place on AxoBase where the big datasets are hosted in a usable form.

Joel: That’s exactly what Xenbase and many other resources are doing. For AxoBase to really be a state-of-the-art modern resource, it has to be tied into, for example, the Alliance of Genome Resources, which links together the model organisms. The other model organism communities have had a great head start (they do have this slight advantage in that their genomes are much smaller). We are in contact with these organizations, and with the NIH, which has comparative genome resources. We want to talk with Ensembl as well. 

All: We are all about aggregation, integration and connection to other organisms. The great challenge with axolotl is it breaks tools because the sequence is so large. We want to make sure that the resources available are linked to the database and connected with known orthologs in other genomes. If we want to apply what we learn in axolotl to other organisms, especially for human health, you have to make sure that the nomenclature and the representations are aligned with each other. Since we are still very much at the formative stages, we want to make sure we don’t waste effort and build AxoBase in ways that are going to have to be reengineered later on to match up with the external model organism community. 

Apart from the four of you, are there any other people who are involved in this resource?

Jessica: The Broad Institute is also involved in this. My main computational collaborator for many years has been Dr. Brian Haas at the Broad Institute here in Cambridge.  He developed the Trinity program for reconstructing de novo transcriptomes, he has a longstanding interest in axolotl biology, and his expertise has been critical in many of our projects. Joel also knows Brian from the olden days, so he’s going to be working with Joel on AxoBase. 

Prayag: Elly Tanaka, Randal Voss and Jeramiah Smith are very crucial in making this happen. We are still figuring out what role they will play in AxoBase. But all of them have been extremely crucial, including Brian Haas, for the success of this portal. I would also like to mention Peter Vize and Aaron Zorn at XenBase, who are very supportive of our effort. The long-term idea is to take a XenBase clone and populate it with the Axolotl datasets.

Joel: XenBase has a model for doing this already. They’ve successfully migrated EchinoBase from a XenBase clone. It makes sense for us to make use of this, in terms of computational efficiency and saving time and money.

All : A lot of people put their money and effort into making progress in the axolotl field. What we are trying to do at AxoBase is not to reinvent the wheel. We want to take everybody along on this journey and give everybody due credit. The main thing we have been doing is to talk with everyone who has been crucial in moving the field forward, whether in axolotl, XenBase or AGR (Alliance of Genome Resources). There is a lot of behind-the-scenes conversation going on with many different parties involved, getting them all on board so that eventually we can have a proper knowledge base like most other established organisms have.

Even though AxoBase is still in its early stages, are there any features or ‘hidden gems’ that you want to highlight? 

Prayag: One of the biggest gems is the antibody collection that James has gathered. If you are working with axolotl, one of the biggest challenges is that most commercially-available antibodies do not work. That’s why on AxoBase we have a list of antibodies that we know work.

James: The antibody collection is community contributed. People can put forward their rockstar antibodies that they are really sure of. They can submit pictures and evidence showing how an antibody works well for a particular application. Hopefully we’ll get more submissions as time goes on.

Where does the funding come from?

Joel: Prayag, James and I are funded through the NIGMS (National Institute of General Medical Sciences), and Jessica’s work is funded by NICHD (National Institute of Child Health and Human Development) and NSF. So you can say the work we’re doing for AxoBase is funded indirectly by the NIH, NICHD and NSF. We are actively seeking other funding, primarily from the NIH, but the NSF is not out of the question. 

Jessica: This is part of the issue. We are all putting in the effort, and at some point, we have to account for it financially. We can’t just scrounge around forever and expect that we’re going to get this awesome resource when people need to be supported. 

What are the plans for AxoBase in the next few months?

Jessica: One feature we want to have on AxoBase is guide RNA prediction for your gene of interest. 

Joel: Another feature that is coming soon is HCR (Hybridization chain reaction) probes. We have predictions for all of the known transcripts, so we just need to build the interface that will allow them to be searched and visualized in a nice way. I’d like to see the HCR probes and the guide RNA prediction available on AxoBase within the next year or so. 

James: HCR is the current standard for in situ hybridization, equivalent to RNA scope used in many mammalian systems. You can multiplex, and it works really well in axolotl to work out where the gene is expressed in your tissue of interest. It’s become an important tool to validate single cell RNA sequencing data.

 Prayag: The other thing is, being a non-standard model organism, we do not have a lot of antibodies. So in situ hybridization techniques are much easier because we know the transcript sequence, and can design the probe against it. This technology is going to be very useful for axolotl researchers. 

Any final words for the Node readers?

AxoBase is an enabling resource. It enables biology to get done. We’re not resequencing anything —we’re putting everything together for improved usability. Our hope is that, not just axolotl people, but even non-axolotl people can easily access all this information as per their need. This will eventually allow more people to use axolotl as a model organism and grow the entire axolotl community.

(No Ratings Yet)

Loading...