Over the past three decades, biological data have grown dramatically in both size and complexity. The major contributors to the growth in size of computation biology data include, but not are not limited to, the ability of biologists to sequence complex genomes such as the human genome (1990–2003) ( Lander et al., 2001), the advent of new high throughput sequencing techniques (around 2008) ( Marx, 2013), and most recently the very rapid advancements in single cell technologies, introduced in 2009 ( Wang & Navin, 2015).

The complexity of biological data has been growing even faster, and doesn’t seem to be linearly dependent on the size of data. Examples of complexity in the field of computational genomics include multiple diverse sources of technical noise, low signal to noise ratio, low numbers of biological replicates in comparative approaches, rare and usually hardly detectable mutations in non-coding regions and rare and barely identifiable cell types in complex heterogeneous systems such as the immune system and/or the brain.

At he intersection of mathematics, statistics and computer science is machine learning (ML), the de facto tool box in data science for deciphering the relationship between the input and output as well as detecting significant patterns within large, complex data sets. These quantitative approaches have been shown to be effective and are becoming increasingly popular in addressing challenges such as those outlined above. Highlights of their successful applications in functional genomics include, but are not limited to, learning and characterizing chromatin states by employing unsupervised approaches such as chromHMM ( Ernst & Kellis, 2012), predicting sequence specificities of DNA- and RNA-binding proteins using convolutional neural networks such as DeepBind ( Alipanahi et al., 2015), and employing a combination of supervised and unsupervised approach to determine the genetic and epigenetic contributors of antibody repertoire diversity ( Bolland et al., 2016). Nowadays it is almost impossible to publish a study on single cell assays without using dimensionality reduction methods such as Principal Component Analysis or t-SNE.

One indirect measure of the success of these techniques in extracting scientific insights from biological data is to measure the popularity and usage of machine learning algorithms in life sciences research over time. I set out to quantify what fraction of published papers in the NCBI database mention a particular technique and how these numbers change over time.

For this analysis, I used the R RISmed package ( Kovalchik, 2015) to parse the publication data from NCBI. I examined publications in PubMed from 1990 to 2017 using a metric that measures the proportion of publications per year that mention the technique in the full text (Hits Per Year per Million articles published, or HPYM). The Popularity Rate (PR) of a technique was then defined as the difference between HPYMs in any two consecutive years. A positive PR shows an increase in popularity, whereas a negative PR reflects a decrease in popularity. I limited this note to 10 models listed in Table 1 which have been the most common or which showed a sharp change in popularity rate at a particular time. However, the R code is available with which any particular model during a specific period of time can be easily measured.

This analysis demonstrates that the overall popularity of machine learning methods in biomedical research has linearly increased since 1990 to 2017, but with two different slopes. From 1990 to 2000 the slope is 0.02, meaning that popularity increased only 2% per year. In 2000 (when sequencing big genomes became possible) the slope increased to 0.06, and since then it has remained constant. Perhaps surprisingly, a maximum of 1.2% of all papers published in PubMed in any calendar year have mentioned one of the machine learning methods investigated in this study ( Figure 1A).

The Linear Regression (LR) models have been the most dominant machine learning techniques in the life sciences over the past three decades ( Figure 1B). It is interesting to see that their popularity rate has not been much effected by the rise of more sophisticated ML techniques such as ensemble-based approaches and/or Support Vector Machines and even with very recent and state of the art deep learning techniques. With a constant increase of 300 HPYM, and considering its higher intercept at 1990, the linear regression models is predicted to be one of the most popular techniques over the next few years.

Perhaps a very surprising observation of this study is the rise and fall of Principle Component Analysis (PCA). PCA became very fashionable between 2000 and 2013. Since then it has been less used less, although it still is the second most popular tool ( Figure 1B).

In early 2000s, unsupervised Hierarchical Clustering alongside newly introduced supervised techniques Support Vector Machines (SVMs) and Random Forests (RFs), showed a sharp rise in usage, which was mainly associated to microarray data analysis. Usage of hierarchical clustering plateaued shortly after its sharp popularity rise in 2000. SVMs kept their popularity longer, for almost a decade in fact, but subsequently dropped to an almost negligible popularity rate. RFs on the other hand, showed less popularity at the beginning of their arrival, but later on (after 2013) they were ranked the second highest in popularity after Deep Neural Networks (DNN) ( Figures 1B and 1C).

During the period between 1990–2017, neural networks have demonstrated considerable fluctuations in popularity. Known as Artificial Neural Networks (ANNs) in the early 1990’s after Linear Regression and PCA, they were the most commonly used techniques until early 2000, when they lost their popularity to MMs, HCs and SVMs and even later to RFs. However, since 2013, when they became known as Deep Neural Networks (DNNs), their usage has increased remarkably, so that they currently have the highest popularity rate ( Figure 1B and 1C).

The dimensionality reduction technique t-distributed Stochastic Neighbour Embedding (t-SNE) published in 2008, has become quickly tailored to all sorts of single cell techniques. It is therefore not surprising to see that t-SNE usage has also been very rapidly growing over the past few years ( Figure 1C).

I have illustrated the rise and fall of ML techniques in life sciences from 1990 to the present day. I chose this period because I believe this is the transition period for life scientists to join the big-data club. With the same R code used in this study to parse the publication data from NCBI, it would be possible to look at any period of time.

It was not very surprising to see LR models as the most commonly used model in the field, since:

a) LR models are one of the oldest ML methods that have been in use in almost any field,

b) Parameters in LR models can be learned by using a training data with just a few data samples.

c) A lot of other models can be placed under this umbrella, for instance by first applying a transformation function.

It was, however, surprising to see the sharp rise and fall of PCA. Perhaps a contributing factor to PCA being the most dominant dimensionality reduction method available in this period was its easy-to-use implementation in R. The question still remains as to why its popularity decreased from 2008 onwards. Perhaps the arrival of more versatile models such as RFs and SVMs which are very capable of handling high dimensionality and dealing with co-linearity in biological data eased the need to use PCA. Additionally, t-SNE as a tremendously growing dimensional reduction model in the field, is establishing itself as a strong competitor for PCA.

ANNs have been fairly popular since the 1990s until around 2004. Around that time more readily useable and less complex techniques became available, such as SVMs, RFs and MMs. However, with the huge investments of giant information companies such as Google leading to very impressive applications of ANNs (now known as DNNs) in various disciplines, their popularity has started to grow again. The sharp increase usage in popularity rate of DNNs over the past few years ( Figure 1C) suggests that DNNs will take the PR lead again in the coming years.

I appreciate that there are limitations to this study. For instance, for the majority of the comparative analyses of gene expression, researchers use a differential expression software and/or package, but cite only the package name and not the underlying statistical or ML technique used in the package. These cases have not been covered in this study. However, this study can be considered an approximation of the extent of machine learning techniques used in life sciences.

In a similar study ( Jensen & Bateman, 2011), Jensen et al. investigated the rise and fall of only a few supervised machine learning techniques in life sciences. This study can be considered an update and extension of Jensen et al’s work, where the search for the mention of a particular technique was limited o the abstracts of papers in PubMed.

Dataset 1: The text file contains the raw data underlying the results presented in this study, i.e. the number of publications in PubMed mentioning each machine learning technique from 1990–2017. These data is further normalized per million for downstream analysis. DOI, 10.5256/f1000research.13016.d184022 ( Koohy, 2017).

R code used to parse the publication data from NCBI is available at: https://github.com/hkoohy/Machine_Learning_in_Life_Sciences

Archived source code as at the time of publication: http://doi.org/10.5281/zenodo.1039642 ( hkoohy, 2017).

License: GNU GENERAL PUBLIC LICENSE