Difference between revisions of "Team:Aalto-Helsinki/Questionnaire"
(headings, discussion, future prospects, acknowledgments) |
(+introduction, results) |
||
Line 76: | Line 76: | ||
<h2>Introduction</h2> | <h2>Introduction</h2> | ||
− | <p></p> | + | <p>Mathematical modeling is a key component of synthetic biology, serving as a crucial link between the idea and realization of an engineered biological system. In our project, modeling played a central role in 1) helping us focus our bioengineering efforts on the rate-limiting steps of our synthetic metabolic propane pathway and 2) giving us useful predictions on whether our concepts could work.</p> |
+ | |||
+ | <p>As is often the case with interdisciplinary work, cooperation between modelers and biologists is not always easy, something we also noticed in our project. If we had understood the potential pitfalls and known good practices of this cooperation, we would have saved a lot of time and been able to achieve more in our project. One could even say that the development of synthetic biology depends on sociology, reflecting the necessity of creating an environment where scientists with different expertises can effectively work as a team (Calvert & Fujimura, 2011).</p> | ||
+ | |||
+ | <p>Whereas interdisciplinary cooperation has previously been studied in professional research environments (Calvert 2010), iGEM offers an unique possibility to study how students perceive and approach the challenges of interdisciplinary work. To specifically study the relationship of mathematical modeling and laboratory work in iGEM teams and gather information about pitfalls, good practices and the integration of modeling and laboratory work we created a questionnaire and sent it to a number of teams. We also studied the expectations students had for modeling, how the modeling efforts were organized and how well modeling and experimentation was integrated in iGEM teams.</p> | ||
+ | |||
+ | <p>Having finished the questionnaire, we found the share of respondents from a mathematical background surprisingly low. To find out how well our sample represents iGEM teams we looked up the majors of students from 84 iGEM teams from 2014. We also compared the results to compositions of professional synthetic biology research groups to see if there are any differences.</p> | ||
<!-- Introduction above --> | <!-- Introduction above --> | ||
Line 85: | Line 91: | ||
<h2>Materials and methods</h2> | <h2>Materials and methods</h2> | ||
− | <p></p> | + | <p>The questionnaire was built in Google Forms and contained the following questions, with each question except number 7 having an open text box for the answer:</p> |
+ | here be the questions | ||
+ | |||
+ | <p>We kept the questionnaire open from 22.7 to 6.8. for a total of 15 days. The questionnaire was first published on our Facebook page and Twitter account. We also advertised the questionnaire during the Nordic iGEM Conference. We contacted 7 iGEM teams through Facebook and 41 teams by e-mail and asked their individual team members to respond to our questionnaire. In total we personally contacted 48 iGEM teams: 22 from Europe, 7 from North America, 4 from Latin America and 15 from Asia.</p> | ||
+ | |||
+ | <h4>iGEM team compositions</h4> | ||
+ | |||
+ | <p>We went through iGEM teams from year 2014 and collected their fields of study. We chose to study approximately one third of the 250 teams registered with geographically proportionate amount of teams. The majors of participants were derived from individual team wikis and grouped under similar fields according to table 1.</p> | ||
+ | |||
+ | here be table 1 | ||
<!-- Materials and methods above --> | <!-- Materials and methods above --> | ||
Line 93: | Line 108: | ||
<h2>Results</h2> | <h2>Results</h2> | ||
+ | <h3>Analysis of the respondents</h3> | ||
+ | <h4>Teams and their geographical location</h4> | ||
+ | <p>We received in total 38 responses, 4 of which came after the deadline. 20 responses came from Europe, 12 from Asia and 3 from Latin America. 3 respondents did not specify their team. The responses came from 16 different teams: 8 from Europe, 6 from Asia and 2 from Latin America.</p> | ||
+ | table 2, responded teams | ||
+ | |||
+ | <h4>Fields of study</h4> | ||
+ | <p>The respondents mainly came from biological sciences. In total we received 39 answers to the field of study -question, with one respondent studying two fields. We grouped the fields identically to the way we grouped the data from studying team compositions of 2014 iGEM teams (see table 1). The share of different categories can be seen in figure 1.</p> | ||
+ | figure 1, fields of study | ||
+ | |||
+ | <h4>Years studied</h4> | ||
+ | <p>Typically a respondent had studied for 3 or 4 years, with these groups representing over half of the respondents. About 25% of respondents had studied for 1 or 2 years and 20% for 5 years or more. More detailed information can be seen in figure 2 (below).</p> | ||
+ | figure 2, years studied | ||
+ | |||
+ | <h4>Modeling organization</h4> | ||
+ | <p>Out of the 16 teams that answered modeling was not a part of the project for only one team, represented by a single respondent. Of the 15 teams that are doing modeling, 12 have a specific group assigned for modeling and 3 teams don’t have one. Results as gathered in figure 3.</p> | ||
+ | Figure 3. Modeling organization in teams. | ||
+ | |||
+ | <h4>Roles of respondents</h4> | ||
+ | <p>In total 14 respondents are taking part in the modeling efforts of their team. The division of fields is shown in figure 4. 23 respondents didn’t mention doing any modeling and one response was unclear. Of the 14 people doing modeling 4 were doing it as their main task. These people were studying physics, physical engineering, computer science and chemistry. Of the 10 people that were involved in modeling, but not as their main task, 7 had background in biotechnology, three in biological sciences and one in management of technology. On the other hand, respondents not working on modeling mostly had bioscience background.</p> | ||
+ | Figure 4. Questionnaire respondents by their fields of study and involvement in modeling. | ||
+ | |||
+ | <h3>Analysis of the answers</h3> | ||
+ | |||
+ | <h4>Expectations for modeling</h4> | ||
+ | “Modeling could help to make your choices within biology more rational.” | ||
+ | - A 4th year life sciences student | ||
+ | |||
+ | <p>Modeling was seen as a way to gain theoretical insight into the biological system: to explore the limits of a system, the space where it can work and its predicted dynamics. Some respondents hoped that modeling would give insight to whether their system could function at all. One respondent mentioned modeling as a way to reach a conclusion if experimentation is not possible. Many respondents expected modeling to give predictions about how the biological system functions.</p> | ||
+ | |||
+ | <p>It was hoped that these predictions would help with focusing the experimental work and thereby minimize the quantity of required experiments. For instance, modeling was seen as a way to explore the bottlenecks and critical parameters of a system. In practical terms it was hoped that it could give guidance on for example which RBS or promoter to choose to make the biological system function as desired.</p> | ||
+ | |||
+ | <p>Some hoped that the modeling could precisely represent the biological system, but more often respondents were rather sceptical of modeling biological systems due to lots of unknown variables and many aspects that are difficult to model.</p> | ||
+ | |||
+ | <h4>Issues in cooperation</h4> | ||
+ | “It is sometimes difficult to communicate, however, more often than not I think collaboration between fields is just an advantage.” | ||
+ | - A 1st year Biomedicine student | ||
+ | |||
+ | <p>Co-operation between modelers and people doing the experiments is a major challenge in applying modeling to synthetic biology. Very many respondents considered communication between these groups difficult. Three reasons to communication problems were mentioned multiple times.</p> | ||
+ | |||
+ | <p>Firstly, as modelers often have a mathematical background and experimentalists a background in life sciences, the knowledge gap makes cooperation difficult. Biologists have a better understanding of how life works, but on the other hand they have a hard time understanding modeling, whereas modelers are in the opposite situation. Especially respondents doing mostly modeling stated that understanding other fields sufficiently takes a lot of time.</p> | ||
+ | |||
+ | <p>Second difficulty mentioned was the different vocabulary used by modelers and experimentalists. For instance, two people may refer to the same experiment or concept with different names, creating confusion and making cooperation more difficult.</p> | ||
+ | |||
+ | <p>A third and according to some respondents the hardest barrier was that modelers and experimentalists have a different way of thinking, a different perspective. </p> | ||
+ | |||
+ | <h4>Solutions to cooperation issues</h4> | ||
+ | “We have been dealing with this by talking with no hesitation, exchanging ideas and experiences and accepting that our knowledge completes each others'” | ||
+ | - A physics student working mainly on modeling | ||
+ | |||
+ | <h5>General meetings</h5> | ||
+ | <p>To resolve the challenges in cooperation between modelers and experimentalists and enhance communication, teams had different methods. In some teams questions and problems of all fields were discussed as a team, which was also seen as a good way of getting all team members to get to better understand different fields. Some teams have weekly meetings where issues are presented and discussed, other teams have meetings each day. Thoroughly explaining models or the biological systems to people not very familiar with the question at hand was considered time-consuming, though.</p> | ||
+ | |||
+ | |||
+ | <h5>Modeling meetings</h5> | ||
+ | <p>Some teams have separate meetings for modeling. In these cases, having experimentalists take part in the modeling meetings and vice versa was seen as a good way to enhance communication. Sometimes instructors also take part in the meetings.</p> | ||
+ | |||
+ | |||
+ | <h5>Working together</h5> | ||
+ | <p>In quite many teams, the cooperation was enhanced and the knowledge gap narrowed by taking the modelers to perform the experiments in together with the biologists, in some cases only for a short time in the beginning of the project. Some teams have people who work both in the laboratory and in the modeling team with full-time modelers to answer their biology-related questions. This was also seen as a way to avoid communication problems between the modelers and the rest of the team. In some teams, the modelers and experimentalists work tightly together. One respondent mentioned that it would have been helpful to have had the experimentalists involved in the modeling right from the beginning.</p> | ||
+ | |||
+ | <h5>One person doing both modeling and experimentation</h5> | ||
+ | <p>In some cases, same people did both experiments and modeling, so there was no need for cooperation at all. Whether this is possible depends on the case: if a single person can tackle the question at hand, cooperation is not necessary. One respondent pointed out however, that while having the same person doing both experiments and modeling reduces communication issues, it can increase the likelihood of errors due to lack of discussion.</p> | ||
+ | |||
+ | <h4>Integration of modeling and experimentation</h4> | ||
+ | “The choice of promoters, RBS sites and plasmids depended greatly on the model data obtained.” | ||
+ | - A 5th year biomedical engineering student | ||
+ | |||
+ | <p>Sufficient integration of modeling and experimental work done in the laboratory was considered difficult but important. About half the teams were able to get modeling results that affected the work done in the laboratory. As often was the expectation, modeling helped some teams in focusing their time and energy on certain designs. For instance, it helped teams in deciding what constructs to create and influenced the choice of promoters, RBS and plasmid backbones. Some teams applied protein degradation according to models they built. One respondent also noted that modeling helped in thinking about the possible influences in the system and in improving experimental setups. Some teams performed experiments to gain parameters for their models and adjusted their models according to experimental results.</p> | ||
+ | |||
+ | <p>When there was no effect or the effect was smaller than desired, it was often a question of time. Building a model takes time, but due to the tight timeframe of the iGEM competition, work in the laboratory needs to start early as well. Therefore, the model might provide results only at a too late stage for it to have any effect in the experiments. Many would have preferred to start the modeling earlier on, had it been possible. One respondent replied that the modeling will not have any significant effect, as their team is doing so little modeling due to lack of expertise. In some cases there was simply not enough time to adjust or validate a model using experimental data, often due to difficulties in performing the experiments successfully. Due to the timing of our questionnaire approximately 6 weeks before the end of the competition, some respondents answered that they do not yet know whether there will be an effect, but were nevertheless mostly optimistic.</p> | ||
+ | |||
+ | <p>In addition to the problems of cooperation, many teams mentioned having problems with finding information required for modeling, for instance different constants. One team had difficulties in finding relevant questions to answer with modeling. To overcome these problems, teams seeked help from outside the team, either from other iGEM teams or from supervisors and advisors. It was also mentioned that having a supervisor specialized in modeling would have been helpful.</p> | ||
+ | |||
+ | <h3>Analysis of the iGEM team compositions</h3> | ||
+ | |||
+ | <h4>General team composition</h4> | ||
+ | |||
+ | <p>We collected data from total of 84 teams with 23 teams from Europe, 28 teams from Asia, 5 teams from Latin America and 28 teams from North America. Our sample consisted of 1063 majors. The majors were categorized under Biology, Biotechnology, Mathematics, Computer Sciences, Physics, Chemistry, Other Engineering and Other majors. All the majors under each category are shown in [table 1].</p> | ||
+ | Figure 6. Fields of iGEM participants by category (n=1063). | ||
+ | |||
+ | <p>The division of the fields is gathered in figure 6. Biggest fields are biology and biotechnology with 48.5% and 26% of students respectively. The fields actively doing modeling according to our survey are a clear minority, with computer science at 6.8% and physics at 2.6% of iGEM participants. Mathematical sciences, mathematics, physics and computer science, make up a total of 11.2% of the fields.</p> | ||
+ | |||
+ | <h4>Composition of synthetic biology research groups</h4> | ||
− | <p></p> | + | <p>Very few research groups have clearly outlined what the exact background of their members are, making it hard to compare with iGEM teams. Systems biology and synthetic biology group in Edinburgh and MIT synthetic biology groups had a total of 83 majors with 24 people left unspecified. Majors were grouped as with the iGEM team composition and are presented in figure 7. About half of the researchers came from biology and the second half from other sciences.</p> |
+ | Figure 7. Fields of synthetic biology researchers by category. (n=84) | ||
<!-- Results above --> | <!-- Results above --> |
Revision as of 18:19, 2 September 2015