Microbelibrary - Annotation of the Bacteriophage 933W Genome: An In-Class Interactive Web-based Exercise

Students should have knowledge of the basics of genetics, the central dogma of molecular biology (DNA to RNA to protein), and an understanding that the entire collection of genes comprise a genome. Extensive computer skills are not required, but the students should be familiar with using a web browser (e.g., Internet Explorer or Mozilla Firefox).

A classroom with wireless Internet access (request that students bring wireless-ready laptops) or a classroom or computer lab with Internet access is required for this exercise. Contact the staff at the ASAP database (pmliss@wisc.edu) and request to set up your own unique version of the bacteriophage 933W genome to annotate with your class. Note: you will be required to submit a username and a password. You and your class will all use the same log in information, so choose a username and password that do not include sensitive information. We recommend using all upper or lower case letters, if possible, to make it easier for a large class to all log in. One suggestion is to use the course name for the username (e.g., BACT650) and the school mascot for the password (e.g., badgers). You can request to set up multiple versions of the genome if using it in multiple classes or during multiple semesters. Note: each will require a new username and password.

Student Version.

Have you ever wondered how biologists make sense of the millions of base pairs of DNA sequence that make up a genome? How are annotations, or the biological information, attached to the sequence to create the files available as a GenBank report at the National Center for Biotechnology Information (NCBI, www.ncbi.nlm.nih.gov)? With the advent of whole genome sequencing, the number of microbial genome sequences available has grown exponentially, yet students are not typically involved in the process of annotating genes and exploring genomic data. When most textbooks address the topic of genomics, they often overlook the steps on how information for annotations is derived. In this exercise, the topic of adding information to newly sequenced genomes is presented. Through the use of bioinformatic tools that query sequence databases such as BLAST and InterProScan, students will examine a predicted open reading frame (ORF) from the bacteriophage 933W genome and add appropriate annotations into the ASAP genome database.

The ASAP database is unique in several ways that facilitate this exercise. An important feature in this system is that every annotation (such as the claim that a particular gene encodes a protein with a specific biological function) has a link to the evidence that supports the assertion made in the annotation. Many databases display the assertion made in the annotation, but do not provide the viewer (or student) with a description of how the information was derived. This feature of ASAP is now highlighted as a model approach to add evidence links for annotations and genes in other genome databases. In other designed modules that use the ASAP database, the link for experimental evidence goes directly to the PubMed abstract, enabling the students to access relevant annotation information quickly and easily. Another reason the ASAP database is useful for this exercise, is that all of the genomes housed in the ASAP database are from the evolutionarily related family the Enterobacteriaceae. Members of this family include Escherichia coli and Salmonella spp., both of which should be familiar terms to students since these organisms are often mentioned in the media in relation to food safety and/or water quality issues. We chose a bacteriophage from E. coli O157:H7, since they are important for the microorganism’s pathogenesis, evolution, and role in public health. A bacteriophage was selected, as opposed to a whole genome of a bacterium, because we wanted to keep the complexity level of the exercise low. A phage genome is smaller and easier to conceptualize than a bacterial genome, which should help students better understand the genome annotation process.

To introduce the topic of annotations, a slide presentation in PowerPoint format is provided to aid the instructor’s presentation of the exercise to the class. The instructor can use these slides to give a brief overview of genomic biology and provide an introduction to the concepts discussed in the exercise described below. We have provided explanations and key topics to address from each slide (see below) in the presentation. All slides are numbered and correspond to the commentary provided.

Slide 1. Introductory slide, with students pictured from an introductory biology course at University of Wisconsin—Madison engaging in this module using wireless ready laptops in a large lecture room equipped with wireless internet. In order to teach topics about genomics, you need to get computers in the students' hands. Based on campus computing survey statistics at UW-Madison in 2007, 77% of undergraduates own their own laptop computer, and laptop checkout programs are prevalent throughout the campus. If you have a small course, it may be feasible to arrange implementation of this module in a computer laboratory setting where students will have access to desktop computers. Alternatively, if your institution is wireless-equipped for the classroom, you can request your students to bring their own laptops. For those students without laptop computers, look into laptop check-out programs at your institution and provide this information to students early in the semester so they may reserve computers for the date that you will conduct this exercise (include in the syllabus). Students can also share a laptop if needed.

Slide 2. Definition of a genome: this slide is meant to serve as an explanation of the basic concept of a genome as the complete complement of genetic material from an organism.

Slide 3. This slide introduces the first genome ever sequenced. The relative small size of phage genomes made them well-suited to be the first genomes to be sequenced completely. This slide mentions one of the pioneers in genome sequencing, Frederick Sanger; through his efforts, the first genome was sequenced of a small phage that contained only 10 genes. Dr. Sanger was later awarded the Nobel Prize in 1980 for this work that started the emergence of the field of genomics.

Slide 4. History of DNA sequencing and genome (continued). This slide mentions the next genome that was sequenced, another phage, but larger than the first. Dr. Sanger and colleagues developed a new method of sequencing called “shotgun” sequencing, which enabled the sequencing of large strands of DNA at a much faster pace (a brief definition of “shotgun” sequencing is provided by Dr. Francis Collins, Former Director of the National Human Genome Research Institute at http://www.genome.gov/glossary.cfm?key=shotgun%20sequencing). This genome contained 46 genes, and a visual layout of the genome containing the genes is provided in the illustration. By now, it should become obvious to the students that the early history of the field of genomics focused on phage, due to their relatively small genome size. Therefore, to teach the subject of annotations to a class, this exercise also uses a phage since it is relatively small (78 genes), and is a genome that a class may be able to completely annotate and sense a feeling of accomplishment.

Slide 5. In 1995, sequencing of the complete genome of Haemophilus influenza was finished, making it the first bacterial genome, as well as the first nonviral genome, to be completed. It was also the largest finished genome at the time, with 1,709 genes. The effort to fully sequence the H. influenza genome took 13 months. When presenting this slide, I use what we call the “JFK analogy.” The day that president John F. Kennedy was shot impacted everyone’s lives, and they always remember where they were when the news was announced. Similarly, when the news of the first bacterial genome sequence completion was announced, an instructor declared that this was a hallmark event for science and one that would surely impact everyone’s lives, as the field of genomics was preparing to explode and would soon map a new avenue of science in microbiology and eventually human medicine.

Slide 6. The first yeast genome is completed. This was the first organism with multiple chromosomes to have its genome sequenced and represented an increase in the genome size (6,269 genes). As stated on the slide, it took an enormous amount of time (7 years) and labor (74 labs) to complete. This information is presented to provide an appreciation of how DNA sequencing technology will advance in the coming decades to the point where it is believed that human genomes will be sequenced in approximately 15 minutes by the year 2013.

Slide 7. Many other completed eukaryotic genomes were soon to follow with completed genome sequences. This slide mentions the next series of hallmark genome projects, with the first fruit fly (13,000 genes), nematode (19,000 genes), and a plant (26,000 genes). This leads to the obvious question and the mission to sequence the largest and most complex genome yet, the human genome.

Slide 8. The first human genome was completed in 2001, and initially it was a race to be credited as the first team to accomplish this hallmark achievement. The teams under the direction of Drs. Craig Venter and Francis Collins joined forces to complete the sequence and share the credit. With the advent of DNA sequencing technology, it is believed that human genomes will be sequenced in less than 15 minutes at a cost of about $100. This means that every student in the classroom will have their own unique genome sequenced by their health care provider, and soon patients will discuss what does this all mean with their physician. The original sequence is thought to contain between 30,000 to 40,000 genes. Why such a discrepancy in the number of genes? The variability involves different results from the process and steps about how genes are predicted. This leads into the next slides that focus on how genes are predicted.

Slide 9. Regardless of the technology used, the outcomes of all genome sequencing projects are long strings of A's, C’s, G’s, and T’s. So how are the genes predicted? At this point ask the students how they would find a gene in the sequence in the slide. What important patterns would they look for? This is the first process of genome annotation, called “structural annotation” or “gene finding/calling.”

Slide 10. Here we define the two main steps involved in genome annotation. For “structural annotation” or “gene finding/gene calling,” it would be too time consuming and laborious to find and predict every gene manually. With the advancements in bioinformatics tools, this process has become largely automated. Yet, even with the best computer-based prediction tools, typically in bacterial genomes, 50 to 90% of the “actual” genes can be found. Some of the first bacterial genomes undergo constant updates to find genes that were missed in initial attempts. This emphasizes that a person(s) will still be required to refine and finish these processes on genomes. Once a gene has been identified, the second step of genome annotation is to attach biological information to the protein that is encoded by the predicted gene.

Slide 11. Some of the patterns or criteria that are involved in structural annotation to identify gene boundaries are that the gene must have start and stop codons. This is often referred to as the open reading frame (ORF). Some of the other patterns utilized in predicting genes include sequences upstream of the predicted ORF. These upstream sequences are typically used as recognition sites in either the DNA or mRNA form (transcription and translation control, respectively).

Slide 12. Once a gene has been identified, how do we figure out what the encoded protein is? This is the process of attaching biological information as annotations to a gene. Some of the types of annotations that can be added to a gene are name, function, product, regulation, molecular interactions, cellular location, etc. As a starting point, focus on the annotations for gene names, products, and functions since the latter two are the types that students will annotate in the exercise. Provided are brief explanations of these annotation types. For further information about each annotation type and the process of adding annotations to the ASAP database, refer to the annotation guide found at https://asap.ahabs.wisc.edu/asap/home.php.

Slide 13. Hypothetical proteins are found in any newly sequenced genome. This slide helps to explain when to annotate a product as a hypothetical protein. This is an important subject that students need to understand—not every gene in a newly sequenced genome will have a known product. Even with the most studied microorganism, E. coli K-12, there are still a large percentage of genes with encoded proteins that we have no idea what they do or the role they play in the cell. The bacteriophage 933W genome will contain genes that will have to be annotated as hypothetical proteins. This presents an area for discussion. At the end of class, take a survey to find out how many students had genes that were annotated as hypothetical proteins. This will allow the students to appreciate that genome annotation is not a simple straightforward procedure.

Slide 14. This is a picture of a GenBank file. Once annotations are added to a gene in a genome, all of this information is required to submit a genome file to the largest database of DNA and genomic sequence data at the National Center for Biotechnology Information. For example, every gene has to have a product annotation. Some of the other informative regions of the GenBank page are highlighted to aid in navigating files at NCBI such as organism, name, product, and function. This is important, because when BLAST searches return results, students will choose sequences that are similar to their respective query and this will lead them to other GenBank files to find information for adding annotations in the exercise.

Slide 15. The largest worldwide database of sequenced DNA and genomes is the National Center for Biotechnology Information (NCBI). Some of the resources that are important when annotating genes are tools such as BLAST and PubMed. BLAST is a powerful tool, because if we find a gene is similar to another characterized gene based on BLAST searching, we can hypothesize that what is true for the first gene, is most likely true for the newly sequenced gene product. PubMed allows searching in the archives of published journal articles for information that aids annotations.

Slide 16. This slide is meant to provide some brief general information about phages and also about the 933W phage genome that students will annotate. The animation illustrates the infection of E. coli by a phage and demonstrates some of the structural components of a phage, and more importantly, how the phage genome is injected and later integrated into the host genome. Some of the genes that students will annotate will encode structural components, replication proteins, and machinery for integration into the host cell. Bacteriophage 933W was found in the first sequenced genome of E. coli O157:H7 strain EDL933 (2).

Slide 17. This is a list of the tools that you and your class will use to annotate the 933W phage genome. For more information about the ERIC genome database, refer to the publication by Glasner et al. (1). The topic of annotations can easily become complex and extensive and there are numerous other tools and resources, but this exercise will focus on two fundamental tools, BLAST and InterProScan. These were selected with the mindset that if you only had 15 minutes to show tools for annotations, select the two that would allow the students to query and learn useful information for the most diverse annotation projects.

Slide 18. Provided are links to additional materials and resources for the ASAP database, NCBI, and Interpro. For simple explanations about using BLAST and InterProScan, refer to the book Bioinformatics for Dummies, chapters 6 (InterProScan) and 7 (BLAST).

The instructor should provide some basic background information about the genome that they are going to analyze. Viruses that infect bacteria are called bacteriophage, or phage for short. Some types of phage can sometimes insert their genome into the genome of a bacterium they infect, thus providing a mechanism for acquisition of new genes by bacteria. Escherichia coli O157:H7, the “bad hamburger” bug, has more than a dozen phage genomes inserted into its genome. One of these, named phage 933W, appears to have carried with it genes for making a nasty molecule called shiga toxin, a protein that causes some of the medical complications in people infected with E. coli O157:H7. In this class exercise, students are given the sequence of the 61,670 base pair 933W phage genome, the predicted boundaries of genes (the structural annotation), and instructions for using web-based sequence analysis tools to examine these genes. The students will employ these tools to infer the biological functions of genes and record their observations as annotations in a genome database.

The ASAP website will serve as the central resource for this portion of the exercise. The genome sequence of phage 933W and its predicted genes are presented to students in the ASAP database. The students are provided with a set of instructions that walk them through the process of investigating the potential function of a particular DNA genome segment and adding their findings to the ASAP database. To facilitate the exercise, the instructor should add annotations to the first coding sequence (CDS) in the genome. CDS regions, also called open reading frames (ORFs), are segments of DNA, usually around 1,000 nucleotides long, that begin with a start codon, end with a stop codon, and encode a potential protein. A primary goal of genome sequencing is often to infer the biological function of the hundreds or thousands of genes encoded by the genome. This typically begins by examining CDS's and comparing their DNA or predicted protein sequences to other sequences with known biological functions. During the demonstration, the instructor should familiarize the students with the steps necessary to access the sequence analysis tools students will use, such as following the links in ERIC to access the BLAST program. The instructor should also explain the BLAST result output format, that shows a list of other proteins from a database ("target" proteins) that match a protein used in a search (the “query” protein). The instructor should teach students how to interpret the scores that indicate the reliability of the matches. BLAST produces two scores for each matching target protein, a bit score and an e-value, the higher the bit score and the lower the e-value, the better the match. Show students the alignments of the sequences that were produced by BLAST and used to compute the scores. The instructor should follow the BLAST exercise with a demonstration of another useful sequence analysis tool called InterProScan. This tool also performs searches of proteins against a database, but tends to focus on matches to families of proteins and smaller portions of proteins that might represent domains or functional sites and may shed light on a particular protein’s function. At this point, the instructor may want to demonstrate searches of the PubMed database with keywords such as protein names or descriptions to identify publications in scientific journals that may contain further functional information about the protein.

Once logged in to the ASAP database and on the Query Genome Annotations page (Slide 4 in student instructions), obtain a list of all 78 CDS's in the 933W genome—select CDS as the feature type in the Basic Options section, at the bottom of the page in the View section type 78 into the area for Records per page, and click the submit bottom. This will provide the list of the 78 feature IDs. You can copy and paste the information into an Excel spreadsheet to subdivide the list of feature IDs for the class based on birthdays. The first CDS in the list is the int gene, which will be used as the example.

When you annotate the first CDS with the class, you should add information for the “product” and explain that the “evidence” that must be provided for each annotation is important, as it describes the method used to arrive at the conclusion.

Annotation type	Value	Evidence type		Reference
Gene	Int	Published annotation (supply GenPept Accession number)		NP_049461
Product	integrase	Published annotation (supply GenPept Accession number)		NP_049461
Function	cleaves DNA substrates	Protein sequence similarity (supply InterPro domain)		IPR002104
Function	site-specific recombination		Published annotation (supply GenPept Accession number)	NP_049461

Common pitfalls encountered.

Some common questions that arose from students and reviewers during field testing are presented, and a response is provided to assist in explanation of the material.

Question 1: What do I do if InterProScan returns no results?

Answer: That means your protein doesn’t have any significant matches in the database. Very likely that means the function is unknown. In this case, add the product annotation as a hypothetical protein. The student's instructions contain the step-by-step guide to add this type of annotation in the ASAP database.

Question 2: When the BLAST results are returned, why are there multiple entries that match at 100% identity with my sequence?

Answer: This is due to numerous entries in Genbank for the same gene, i.e., when the genome of the phage was sequenced (3), when the host genome for E. coli O157:H7 EDL933 was sequenced (2), and additionally, there has been another strain of E. coli O157:H7 (Sakai) sequenced to completion. So there are Genbank entries for the phage genome and the whole bacterial genome. It is likely that the BLAST search will return these four matches with 100% identity. Any of the four files are appropriate for adding as the evidence for annotations, it is up to the students to decide which to use. Students may also be interested in looking at whether there are any substantive differences in the annotation of those redundant entries.

Question 3: Is the first hit on the BLAST or InterProScan list the right one? How do you make the call of one function versus another?

Answer: This question reveals an assumption that many students make, that there is only one correct answer when annotating. This module is designed to show that there are numerous tools and methods for finding information to add biological annotations, and there is a degree of uncertainty in basing annotations on matches in databases. The goal of turning students into annotators is that they learn to use these tools and engage in the decision-making processes required to analyze genomic sequences.

Question 4: Do students ever disagree on an annotation? How is this resolved?

Answer: In this module, each student annotates his or her own gene. In the exception that more than 78 students are in a class or have similar birthdays, there will be multiple students adding annotations to the same gene. It has been our experience that the level of matching similarity from the BLAST search results does not lead to much discrepancy. The database permits addition of multiple entries of annotations, so if multiple students add products and functions, that is fine. The disagreements are what make the process interesting.

Question 5: Will students see the curator approval status—that is, rejected annotations?

Answer: This exercise is set up so that the genome that the class is working on is a “private” genome, which means that only the instructor and the class will be able to see the associated annotations. Further, since this is a previously published genome (3), it does not have to go through the rigorous steps involved in curation of the genome for GenBank upload. This module is really designed to teach the students about how to use the tools and actually go about the process of annotating a gene of a genome. To teach a class on annotation of an newly sequenced genome would require much more time, and it is not within the scope of the learning objectives of this module.

Question 6: How long is one class period?

Answer: During field testing, each class period was 50 minutes. We feel that the exercise can be accomplished in this amount of time. If the class does not finish, students can finish the annotations on their own, and the next class period can be used for discussion and follow-up activities.

This exercise has been used twice at the University of Wisconsin—Madison, the first time in an introductory biology course with 165 students (Biology 151) and the second in an upper-level microbiology course with 15 students (Bacteriology 650: Advanced Foodborne Pathogens). Since the initial review, this exercise was used a third time with an upper-level microbiology lab course with 12 students at the University of Wisconsin—Milwaukee (Biology 580: Experimental Microbiology). A screen shot of a correctly annotated gene generated by a student is provided in Fig. 1.

FIG. 1. Screen shot of a correctly annotated gene generated by a student in Biology 580 at UW—Milwaukee.

Formal Assessment Methods.

For the students in Biology 151 and Bacteriology 650, the first set of pre- and posttest questions was used to assess student learning. The following questions were posed to the class, and results are provided. Following reviewer’s suggestions, many of the questions were rewritten and used to assess learning in an upper-level microbiology lab course at UW—Milwaukee. Overall, the expectations for an introductory biology course are quite different from those for an upper-level course composed of microbiology majors. Therefore, we have provided the results from both scenarios and conducted statistical analysis using the chi-square test when applicable. Additionally, to test if learning objectives were achieved, the percentage of correct product and function annotations were assessed from the microbiology lab course at UW—Milwaukee. It was determined that 83% and 89% of the annotations were added correctly for product and function annotations, respectively.

Pre- and Posttest questions set #1

2) Which of these web-based resources is useful to find biological information about a gene sequence?

3) Annotation of a gene in a genome database includes the following steps, except:

D) Conducting experiments to confirm information obtained through database searches

1. Within a sequenced microbial genome, a gene predicted to encode a protein should contain which of the following characteristics?

A) a region of DNA upstream that corresponds to the promoter binding of RNA polymerase

C) a region of sequence that corresponds to a stop codon (TGA, TAG, or TAA) in the same reading frame as the start codon

D) all of the above

2. What percentage of the protein coding genes do you think automated computer approaches applied to a newly sequenced microbial genome will find?

B) more than 50% of the actual protein coding genes

3. What type of biological annotation cannot be assigned to a newly sequenced gene based solely on comparisons to known proteins or genes?

C) three-dimensional structure

4. In a newly sequenced microbial genome, every identified gene produces a protein that is similar to a known protein? True or false?

5. Which of these web-based resources are useful to find biological information about a gene sequence?

“I really enjoyed learning more about bacterial genetics and the tools that are available online for genomic research and gene identification. This is an area of bacteriology that I have little experience in and I think that having experience using these websites will prove valuable as my research continues.” –UW—Madison student in Bacteriology 650

“The concepts of using BLAST and InterProScan are pretty neat, and it is great that anyone can access this information, not just the insider scientists that put it together. Thank you for teaching our class how to use these tools! I doubt I would have ever learned this stuff on my own had you not taught us.” –UW—Madison student in Bacteriology 650

“It's really helpful to annotate genes, submit references into function and add new interactions for known proteins. It's also great to have a hands-on project instead of learning all of it in theory. I think it's really important to make sure all annotators do follow the same standard, so that product and function annotations are appropriate.”

–UW—Milwaukee student in Biology 580

Pictures of Introductory Biology students at UW—Madison annotating the bacteriophage 933W genome using wireless-ready laptop computers in a lecture hall equipped with wireless Internet.

Once students have learned what annotations are and how the information is derived, and are familiar with using the ERIC database, inquiries into any of the other Enterobacteria genomes such as Salmonella spp., E. coli, Yersinia spp., Shigella spp., or various plant pathogens are possible. Since all members of the Enterobacteria are thought to have descended from a common ancestor, concepts can apply to more advanced topics including evolution, biochemical pathways, gene-protein relationships, or comparative genomics. Additionally, more modules have been designed that utilize the ERIC database and the topic of comparative genomics for inquiry-based learning and were presented at the ASM Conference for Undergraduate Educators 2008 meeting. The slides are available for an overview (The Genomics Era: A Vast Resource for Educators by Dave Baumler http://www.asmcue.org/page02d.shtml) and contain information to integrate these more advanced topics into your course(s).

An alternative approach for using this exercise in a smaller class is to assign more than one CDS to each student. If the students can’t finish the annotations of multiple CDS's in one class, they can finish the work at home, and the class discussion can occur during the next class period. As a result of annotating the whole genome as a class, the discussion will be more interesting since the students will be able to analyze the genome in its entirety.

This exercise, if used with an upper-level class, can be extended with an additional individual project for students based on the annotation of the genome. The upper-level students could be asked to write a report or prepare a presentation. Some questions to ask include:

2) Did your annotation agree with others who had the same CDS? If not, explain why this may have happened and which annotation appears to be more accurate?

1. Glasner, J. D., M. Rusch, P. Liss, G. Plunkett III, E. L. Cabot, A. Darling, B. D. Anderson, P. Infield-Harm, M. C. Gilson, and N. T. Perna. 2006. ASAP: a resource for annotating, curating, comparing, and disseminating genomic data. Nucleic Acids Res. 34: D41-D45.

2. Perna, N. T., G. Plunkett III, V. Burland, B. Mau, J. D. Glasner, D. J. Rose, G. F. Mayhew, P. S. Evans, J. Gregor, H. A. Kirkpatrick, G. Pósfai, J. Hackett, S. Klink, A. Boutin, Y. Shao, L. Miller, E. J. Grotbeck, N. W. Davis, A. Lim, E. T. Dimalanta, K. D. Potamousis, J. Apodaca, T. S. Anantharaman, J. Lin, G. Yen, D. C. Schwartz, R. A. Welch, and F. R. Blattner. 2001. Genome sequence of enterohaemorrhagic Escherichia coli O157:H7. Nature 409:529–533.

3. Plunkett III, G., D. J. Rose, T. J. Durfee, and F. R. Blattner. 1999. Sequence of shiga toxin 2 phage 933W from Escherichia coli O157:H7: shiga toxin as a phage late-gene product. J. Bacteriol. 181:1767–1778.