Several people above have recommended ribosomal genes of one sort or another. i would strongly recommend against these, based on the difficulty of alignment/homology determination of length variable regions. These are mainly used by people interested in deeper-level relationships, and by people who want to mine a lot of homologous sequences out of genbank. I suspect that if molecular systematics started over today, very few systematists would use these difficult-to-align genes.

Protein-coding genes - either mitochondrial or nuclear - have the advantage of being translatable, and at the level of species and genera usually do not contain a lot of length-variable regions (if you avoid introns, anyway). Even for "conserved" genes, the third positions are mostly neutral and therefore not constrained by selection, and so at least 33% of the characters are free to vary.

Probably your best bet to find some useful protein-coding genes would be to look in genbank and see what genes have been used by other nematode workers.

If there are several mitochondrial or nuclear genomes close to the group she is interested, she could try to select automatically the best marker genes: using some existing methods such as:

http://www.ncbi.nlm.nih.gov/pubmed/22474405

Hi. It can not be determined a priori, because the rate of evolution of the same genes in different taxa of animals can vary greatly. Your colleague should study the literature specifically on nematodes. Usualiy COI is suitable for such work, and, if they need more variability, the ITS. But it is only in general.

I would claim to avoid generalizations. The best way to approach, for me, is sequencing different common genetic markers (let´s say, 28S rRNA, COI and ITS) and compare different well-morphologically-established species. From this information you can establish your own thresholds among species or populations (or whatever taxonomic level you are studying). I work with tardigrades, and we have found them extremely high variable for genes commonly used for DNA barcoding (which, trust me, is not so uncommon for certain organisms), with the associated problem of substitution saturation. For example, tardigrades have a interespecific variation for the COI from 15 to 17%. I am studying a fragment of the 28S rRNA which seems to be useful to discriminate among species, at least in one of the Tardigrada classes. And I have found significant differences among species with the 18S rRNA. So everything is relative to the organism you are studying.

for species level phylogenies you can use a fast-evolving gene, such as the COI.

for deeper phylogenies (family level) the COI is not so useful because it tends to reach saturation and is not informative enough (specially third codon positions). In this case slow evolving genes from the nuclear genome can help.

The best way is sequencing different common genetic markers such as 28S rRNA and ITS

to supplement Carlos, in insects (Lepidoptera) informative genes that seem to work fairly well from generic to family level seem especially to be CAD, Ef1a and Wingless, to start with. But, as already emphasized by others, their applicability may be very taxon specific, and if there exists any relevant nematodean literature, I would carefully check what is said there. The benefit of nuclear genes is their often straightforward alignment, as the gene sequences tend to have few insertions and deletions, to generalize, while the ribosomal ones may be quite the opposite.

Hi Yedid,

These articles would suit your need.

The nuclear encoded gene of second internal transcribed spacer (ITS2) had been widely used in recent year and had been found useful especially in discerning cryptic and pseudo-cryptic species; not only in higher plant, insects but also microalgae. This region is relatively short and can be easily sequence through PCR. Hope this would help.

Annette W. Coleman

ITS2 is a double-edged tool for eukaryote evolutionary comparisons

Trends in Genetics. (2003) 19(7):370-375

Annette W. Coleman

Pan-eukaryote ITS2 homologies revealed by RNA secondary structure

Nucl. Acids Res. (2007) 35(10): 3322-3329

Jörg Schultz, Tobias Müller, Marco Achtziger, Philipp N. Seibel, Thomas Dandekar, and Matthias Wolf

The internal transcribed spacer 2 database—a web server for (not only) low level phylogenetic analyses

Nucl. Acids Res. 34(suppl 2): W704-W707

Lim, H.C., Leaw, C.P., Su, S. N.P., Teng, S.T., Usup, G., Mohammad-Noor, N., Lundholm, N., Kotaki, Y. and Lim, P.T. (2012), MORPHOLOGY AND MOLECULAR CHARACTERIZATION OF PSEUDO-NITZSCHIA (BACILLARIOPHYCEAE) FROM MALAYSIAN BORNEO, INCLUDING THE NEW SPECIES PSEUDO-NITZSCHIA CIRCUMPORA SP. NOV. Journal of Phycology, 48: 1232–1247. doi: 10.1111/j.1529-8817.2012.01213.x

Christian Koetschan, Frank Förster, Alexander Keller, Tina Schleicher, Benjamin Ruderisch, Roland Schwarz, Tobias Müller, Matthias Wolf, and Jörg Schultz

The ITS2 Database III—sequences and structures for phylogeny

Nucl. Acids Res. (2010) 38(suppl 1): D275-D279

As many as you can and - if the organisms under study allow for it - with various transmission modes.

I think you should notice that one theme runs through all of the answers so far: There is no hard and fast rule about which molecular markers to use for relationships at the species or genus level because of different rates of evolution in different lineages. Investigations of a taxon with few or no existing sequences in any database will usually start with sequencing the 18S gene because this region can be amplified with universal primers and there will usually be sequences of related taxa in the commonly used databases. Trees are only the beginning of such an analysis, though. While still dealing with just 18S sequences, you can discover much about relationships by looking at the alignment to discover molecular characters that are unique to your clades of interest. These become a means of assessing the probability that relationships exist when support for tree nodes might still be ambiguous. Looking at secondary structure of the hypervariable regions of the 18S gene can be similarly instructive. Like some of the answers above, I also like using the ITS2 region for assessing relationships at the level of species in my group (ciliates). The only disadvantage is that it may be difficult to find ITS2 sequences of related taxa in databases. This region is not as universally sequenced as the 18S. Secondary structure of ITS regions has been demonstrated to be a powerful clue to relationships at the species/population level (see the papers by Coleman cited above).

It will be depend on the deep of your phylogenetic analysis. As you commented you will need different markers if you are working at species level or in a higher scale as genera, class, etc. For species level it will be interesting analyze genes with higher rates of evolution (as some mitochondrial genes). But you must be careful and be sure that no recombinations are present in your genes, you can check it using RDP3 or GARD method implemented in Datamonkey.

In addition, I recommend to use 3 or more markers and combine them in supernetworks or supertrees.

28S rRNA sequences are really useful for separating genera of a fungal family. I am not sure if it works in nematodes.

I am going to self promote here.

Please, take a look to the following papers.

PhyDesign: an online application for profiling phylogenetic informativeness.

http://www.biomedcentral.com/1471-2148/11/152

Phylogenetic signal and noise: predicting the power of a data set to resolve phylogeny.

http://sysbio.oxfordjournals.org/content/61/5/835.abstract

Several people above have recommended ribosomal genes of one sort or another. i would strongly recommend against these, based on the difficulty of alignment/homology determination of length variable regions. These are mainly used by people interested in deeper-level relationships, and by people who want to mine a lot of homologous sequences out of genbank. I suspect that if molecular systematics started over today, very few systematists would use these difficult-to-align genes.

Protein-coding genes - either mitochondrial or nuclear - have the advantage of being translatable, and at the level of species and genera usually do not contain a lot of length-variable regions (if you avoid introns, anyway). Even for "conserved" genes, the third positions are mostly neutral and therefore not constrained by selection, and so at least 33% of the characters are free to vary.

Probably your best bet to find some useful protein-coding genes would be to look in genbank and see what genes have been used by other nematode workers.

I strongly recommend 16s, 18s and Cox 1 genes belonging to mitochondrial DNA because they have conserved regions that allow to do phylogenetic analysis

I worked with Cox1 and i had very accurate data.

Take a look to this paper:

Acanthocheilonema delicata n. sp. (Nematoda: Filarioidea) from Japanese badgers (Meles anakuma): Description, molecular identification, and Wolbachia screening.

Uni S, Bain O, Suzuki K, Agatsuma T, Harada M, Motokawa M, Martin C, Lefoulon E, Fukuda M, Takaoka H.

Source

Institute of Biological Sciences, Faculty of Science, University of Malaya, 50603 Kuala Lumpur, Malaysia; Department of Parasitology, Graduate School of Medicine, Osaka City University, Abeno-ku, Osaka 545-8585, Japan. Electronic address: [email protected].

http://www.sciencedirect.com/science/article/pii/S1383576912001055#

If you can't download the paper, tell me and i will send you a copy.

Hi,

as most of the people before me said that an 18S rRNA data approach has its problems: I understand that this is - generally spoken - due to the general fact that this is a phylogenetic approach. Such problems you would have also with any other single-gene (protein coding) gene family due to the heterotachy of gene families. Maybe a spacer approach is more successful (no experience). The point however is that people try to circumvent the one-gene-family approach by using phylogenomics. This can be multiply solved. Either you are using a set of (e.g. housekeeping) genes and concatenate them on sequence level and make a multiple sequence alignment (MSA) and make a tree, etc. Or you are using generating a super tree from several gene (family) trees. Or you are producing a gene content tree including the full sets of gene families on different levels which have their implications, respectively. As I understand a tree based on MSA of concatenated genes is the state-of-the-art method.

Maybe the correct answer is dependent also from the taxonomic level you are interested in. Do you wish to investigate a particular, close subgroup, an order, a family or do you want to make a full species approach?

We did a comparison about the topic as it can be read at:

http://dx.doi.org/10.4137/EBO.S9642

Welcome!

COX I gene

I used 12s, 16s y CO1 to clarify the phylogenetic relations between species of a genre of insects, it were very heplful. I also try to use ITS but they didnt work for me. the best thing you can do its to look some text or articles related with your group.

Hi Diana,

Using the whole ITS1-5.8S-ITS2 might not help to resolve the phylogeny. As ITS1 was way too diverse. However if you work out more on the ITS2 alone together with the secondary structure information, i'm sure that you would get what you want.

The ITS2 secondary structures is found to be conserved that in general they'll form four main helices (helix I-IV) when you fold it. You can try using RNAStructure (http://rna.urmc.rochester.edu/software.html) or you can look for the template through literature. Followed by homologous modeling using ITS2 Database III (http://its2.bioapps.biozentrum.uni-wuerzburg.de/).

This is a paper discussing about blue butterfly:

Martin Wiemers*, Alexander Keller and Matthias Wolf

ITS2 secondary structure improves phylogeny estimation in a radiation of blue butterflies of the subgenus Agrodiaetus (Lepidoptera: Lycaenidae: Polyommatus )

BMC Evolutionary Biology 2009, 9:300

You can also look into this paper for all the analyses steps:

Jörg Schultz, Matthias Wolf

ITS2 sequence–structure analysis in phylogenetics: A how-to manual for molecular systematics

MPE 52(2): 520–523

Other than align the sequences alone you will also incorporate it with secondary structure information and it actually helped to improve the alignment. Nonetheless the compensatory base change (CBC) and hemi-CBC that had been used as another evidence on sexual incompatibility. Our group had been worked on harmful algae, marine diatom Pseudo-nitzschia and information from ITS2 had always been useful. We actually described new species with using ITS2 as supportive evidences. It's also useful in design DNA barcoding.

Dear Gabriel.

Better to use combinations of 2 or more regions. Most people use ITS, ETS, atpB, 18S rRNA. Combinations gives better and more reliable trees.

For resolving at species or genus level, other marker systems like EST SSRs, AFLPs, SNPs are also as effective.

I do not know well nematode species, but I know genus-level phylogentic analysis in insect using combined mtDNA gene marker (16S mitochondrial genes) and nuclear gene marker.

Hi.

I think the best is that anyone constructs their own database for this decision. The same gene can have very distinct evolutionary patterns in diferent organisms and this pattern could be different in some levels of their evolutionary tree... You can use the previosuly bibliography and the data from Genbank to make your database and to have a closer knowledge of how works each gene in the taxonomic level that you are interested.

Good luck!

Hi,

A combination of more than five nuclear and mitochondrial genes are advisable for robust phylogeny. If you have more genes, better the output. Also, I suggest you to select those genes for your target taxa which (sequences) are already available in GenBank (NCBI etc. ) from other's publication so that in future It will be feasible to compare taxa across the region. Keep your objectives and budget (expense) in your mind when selecting your genes.

Hi Gabriel,

There are some things your friend might need to consider for her study, for example:

1- What is the nematode group she is working with? Are there primers already available for this group?

2- Is she willing to amplify the same genes for all her taxa of interest or is she thinking to use a lot of the information already available on GenBank?

These two questions are key aspects because it will basically determine how much work your friend will have to do.

Most of the DNA sequences for nematodes (free living and parasitic species) are based on ribosomal genes, especially the 18S gene. There is a large database already for nematodes based on this gene and that is why people use it when the focus of the study is phylogeny! In my experience, working with some of the genes for nematodes, here is what I have noticed:

18S: about 2000 bp, quite conserve, and as you already know it is not going to give you any resolution at the species/genus or even family level. However, if the phylogeny is looking for both deeper and more recent relationships, it can be useful.

28S: most D2D3 domains (about 800 bp), some papers are also based on the D1-D3 domains, including D2 (about 1600 bp). It is more variable compared with 18S; it can be much more useful for recent phylogenetic relationships (perhaps genus and family levels, sometimes species).

ITS (including ITS-1, 5.8S, ITS2, about 1000 bp): very variable region, the alignment can be a little trick, it has been used for some population genetic studies on marine and parasitic nematodes. In fact, this gene can be very variable within a species/population, so it can produce more than a unique sequence/haplotype when multiple individuals are PCR/sequenced.

Besides ribosomal genes:

COI: the database is still limited, mostly due to the lack of universal primers for nematodes. However, there are some sequences already available on GenBank. It has been also used in population genetic studies for nematodes.

Hsp90: it has been used more restricted to plant parasitic nematodes.

I hope this information can be helpful!

Best,

Tiago

A heartfelt thanks to everybody who responded! Thanks everyone for your valuable input!

it depend on the taxa you want to work. For example, in Diplogasterid nematoodes 18S rDNA, or in Predatory nematodes (Mononchida) 18S rDNA make a better phylogenetic results. Finally you can also check the database of NCBI, and make a decision, but for desired results if you choose at least two genes (e.g. mtDNA and rDNA) you may obtain better result.

Have a luck

I think that ribosomal genes (as mentioned by A. brower above) are very difficult. I worked a lot with them and clearly you need to rebuild the secondary structure to have a chance to perform alignment.

You've got to know if your species are close or not.

For example Cox1 (the classical barcoding gene) is ok if species are well defined but is a catastrophic tool between close species.

There is an old review but very useful on mitochondrial genes (on insects but you can apply to your nematodes):

Simon, C., Frati, F., Beckenbach, A., Crespi, B., Liu, H., and Flook, P.

(1994). Evolution, weighting and phylogenetic utility of mitochondrial

gene sequences and a compilation of conserved polymerase

chain reaction primers. Ann. Entomol. Soc. Am. 87: 651–701.

There is a comparison of resolution of the various genes and it is very helpful.

Cheers from Luc

Hi,

We have recently investigated this issue and have proposed a strategy to use available genomes (from related species) to select the best possible combination of marker genes of a minimal size.

You can read the approach and access all the necessary tools here:

A phylogenomics approach for selecting robust sets of phylogenetic markers.

http://www.ncbi.nlm.nih.gov/pubmed/24476915

Hope is useful.

Toni

- First of all, the question was asked 1,5 y ago. Nevermind, acuality remains.

- Then, consider mistakes in the relevant databases regarding single gene families like ribosomal genes:

Ashelford et al., Appl Environ Microbiol 71, 7724-36: At least 1 in 20 16S rRNA sequence records currently held in public repositories is estimated to contain substantial anomalies

Ashelford et al., Appl Environ Microbiol 72, 5734-41: New screening software shows that most recent large 16S rRNA gene clone libraries contain chimeras

- Therefore, people generate phylogenomic trees combining severel genes together, e.g. housekeeping genes. However they do (supertrees, multiple sequence alignments, gene content trees).

For social wasps (Vespidae), we amplified three mitochondrial (16S, cytochrome b, and cytochrome c oxidase I) and one nuclear (wingless) gene, for phylogenetic analysis at genus/species level.