Here are five different ways to gain an introduction to the topic of CNNs. Each approach is geared toward a different style of learning:
Visualize them in real time with your own inputs (this is amazing!)
Watch a lecture by the “godfather” of neural nets, Geoff Hinton.
Take a top-ranked online course on Deep Learning.
Learn the math behind them.
Are differentially expressed (DE) genes also phenotypically important?
A new paper in Cell Reports utilizes RNA-seq and Tn-seq (the “tn” in tn-seq stands for transposon) to map the transcriptional and fitness changes in bacterial gene networks in response to stressors, like nutrient depletion and antibiotics.
The transcriptional response measures changes in gene expression as measured by RNA-seq. The fitness or phenotypic response describes the importance of each gene to the response. This is measured by a different assay, Tn-seq, which takes advantage of transposon insertion to selectively inactivate genes in the bacterial genome. Those genes that are depleted in the stressor condition are determined to be “high fitness” (owing to the fact that the bacteria without those genes died under stress).
First, before even considering DE genes, they found that there is no correlation between a gene’s transcriptional abundance (not fold change) and it’s fitness. While most high-fitness genes were also high abundance, many more high-abundance genes were not high-fitness. Thus, there is no useful relationship between a gene’s abundance and fitness.
Superficially, however, one might expect that genes that show large changes in abundance (i.e., large DE) in response to stressors would also be critical for the phenotypic expression of the bacteria’s stress response. That is, those genes with high differential expression would confer high fitness on the cell.
Testing the DE / high fitness relationship
As it turns out, little is actually known about this, and in this paper, Opijnen, et. al., set out to test the idea to determine if in fact high DE genes are also high fitness.
The researchers looked at comparing differential expression in response to a reduced nutrient environment (a type of minimal media) and an antibiotic stress versus high-fitness genes. They found no correlation:
You can see from the figure that high fitness genes (those on the far left of the x-axis), are not correlated to high DE genes. There are no genes in the upper-left quadrant of either plot, showing that there is no correlation between fitness and high DE in response to either nutrient or antibiotic stress.
Gene networks co-localize high DE and high fitness genes
Even though the authors found no correlation between DE and fitness changes for individual genes, they took the next step and constructed a metabolic gene network for the S. pneumoniae bacteria. Mapping the DE and fitness changes onto this network revealed a key finding: the high DE genes co-localize in pathways with the high fitness genes. That is, a biochemical pathway might have some members that are high DE, and others that are high fitness. An example of this is the shikimate pathway shown below:
The first half of the pathway consists of six genes with significant fitness changes (red boxes) in a row. The next seven genes, from the Trp branchpoint (blue dashed line) are not high-fitness, but do show high DE expression, with four reaching statistical significance. It is not really understood why this happens, but the authors theorize that having the bottom half of the pathway under transcriptional control allows the bacterial to control flux into Trp synthesis and other AA sub-pathways while always maintaining a stable supply of the starting point intermediates (the product of SP1374) through reversible, end product-regulated biosynthesis.
Transcriptomic data should not be used as a surrogate for functional importance
The authors point out that the reliance on trancriptonal abundance changes as markers for functional importance in bacteria, particularly in drug discovery efforts, may be misguided and need to be revisited in light of this and other studies. They also point on that the response to an “orderly” stressor (like nutrient depletion) for which the bacterium is evolved, is likely to be much more clearly defined on a network basis. While the response to a disorderly stressor (a novel antibiotic, for example) may provoke a disorderly transcriptional and fitnress response that can’t easily be interpreted from network analysis. This has important implications for the design of next-generation antibiotics.
Integrative genomics sheds new light on metastatic cancer
A new study from the University of Michigan Comprehensive Cancer Center has just been released that represents an in-depth look at the genomics of metastatic cancer, as opposed to primary tumors. This work involved DNA- and RNA-Seq of solid metastatic tumors of 500 adult patients, as well as matched normal tissue sequencing for detection of somatic vs. germline variants.
A good overview of the study at the level of scientific layperson can be found in this press release. It summarizes the key findings (many of which are striking and novel):
- A significant increase in mutational burden of metastatic tumors vs. primary tumors.
- A long-tailed distribution of mutational frequencies (i.e., few genes were mutated at a high rate, yet many genes were mutated).
- About twelve percent of patients harbored germline variants that are suspected to predispose to cancer and metastasis, and 75% of those variants were in DNA repair pathways.
- Across the cohort, 37% of patient tumors harbored gene fusions that either drove metastasis or suppressed the cells anti-tumor functions.
- RNA-Seq showed that metastatic tumors are significantly de-differentiated, and fall into two classes: proliferative and EMT-like (endothelial-to-mesenchymal transition).
A brief look at the data
This study provides a high-level view onto the mutational burden of metastatic cancer vis-a-vis primary tumors. Figure 1C from the paper shows the comparison of mutation rates in different tumor types in the TCGA (The Cancer Genome Atlas) primary tumors and the MET500 (metastatic cohort).
Here we can see that in most cases (colored bars), metastatic cancers had statistically significant increases in mutational rates. The figure shows that tumors with low mutational rates “sped up” a lot as compared with those primary tumor types that already had high rates.
Supplemental Figure 1d (below) shows how often key tumor suppressor and oncogenes are altered in metastatic cancer vs. primary tumors. TP53 is found to be altered more frequently in metastatic thyroid, colon, lung, prostate, breast, and bladder cancers. PTEN is mutated more in prostate tumors. GNAS and PIK3CA are mutated more in thymoma, although this finding doesn’t reach significance in this case. KRAS is altered more in colon and esophagus cancers, but again, these findings don’t reach significance after multiple correction.
One other figure I’d like to highlight briefly is Figure 3C from the paper, shown below:
I wanted to mention this figure to illustrate the terrifying complexity of cancer. Knowing which oncogenes are mutated, in which positions, and the effects of those mutations on gene expression networks is not enough to understand tumor evolution and metastasis. There are also new genes being created that do totally new things, and these are unique on a per tumor basis. None of the above structures have ever been observed before, and yet they were all seen from a survey of just 500 cancers. In fact, ~40% of the tumors in the study cohort harbored at least one fusion suspected to be pathogenic.
There is much more to this work, but I will leave it to interested readers to go read the entire study. I think this work is obviously tremendously important and novel, and represents the future of personalized medicine. That is, a patient undergoing treatment for cancer will have their tumor or tumors biopsied and sequenced cumulatively over time to understand how the disease has evolved and is evolving, and to ascertain what weaknesses can be exploited for successful treatment.
This blog named a “Top 75 in Bioinformatics” by Feedspot.com!
I made the list at #58. I’m proud of that fact, but I want to push into the top 30 on the internet. I plan to increase my rate of posting new articles and also up my game on content and analysis. Stay tuned!
My collaborators and I just published our study of freshwater mussels and sediment biology using metagenomic methods.
Please check it out!
Kallisto and sleuth are recently developed tools for the quantitation and statistical analysis of RNA-Seq data. The tools are fast and accurate, relying on pseudoalignment concepts rather than traditional alignment. They seem to be gaining popularity owing to ease of use and speed that makes them accessible to users on a laptop.
One thing that has been lacking is proper documentation of these tools. This appears to be changing as more tutorials and walkthroughs become available in the past few months.
I wanted to aggregate some of those here for my own reference and also to help others who may be looking for guidance.
kallisto (rapid RNA-Seq read quantification)
sleuth (statistical modeling and analysis)
sleuth tutorial blog posts:
I just got back from Great Lakes Bio 2017 (GLBIO2017) at the University of Illinois-Chicago (UIC) campus. It was a great meeting and I really enjoyed the quality of the research presented as well as the atmosphere of the campus and neighborhood.
I was very surprised by just how nice the Chicago “West Loop” neighborhood near Randolph Street and down towards Greektown really is. I had some great meals, including a memorable Italian dinner at Formentos.
But the purpose of this post is to briefly describe a few of my favorite talks from the meeting. So here goes, in no particular order:
Kevin White, Tempus Labs:
I was really impressed with Kevin White’s GLBIO2017 talk and demo of his company’s technology (despite the ongoing technical A/V issues!) Tempus labs is a clinical sequencing company but also an informatics company focused on cancer treatment that seeks to pull together all of the disparate pieces of patient data that float around in EHR databases and are oftentimes not connected in meaningful ways.
The company sequences patient samples (whole exome and whole genome) and then also hoovers up reams of patient EHR data using Optical Character Recognition (OCR), Natural Language Processing (NLP), and human expert curation to turn the free-form flat text of medical records from different clinics and systems into a form of “tidy data” that can be accessed from an internal database.
Then, clinical and genomic data are combined for each patient in a deep-learning system that looks at treatments and outcomes for other similar patients and presents the clinician with charts that show how patients in similar circumstances fared with varying treatments, given certain facts of genotype and tumor progression, etc… The system is pitched as “decision support” rather than artificial “decision making.” That is, a human doctor is still the primary decider of treatment for each patient, but the Tempus deep learning system will provide expert support and suggest probabilities for success at each critical care decision point.
The system also learns and identifies ongoing clinical trials, and will present relevant trials to the clinician so that patients can be informed of possibly beneficial trials that they can join.
Murat Eren, merenlab.org
Murat Eren’s talk on tracking microbial colonization in fecal microbiome transplantation (i.e., “poop pills”) was excellent and very exciting. Although the “n” was small (just 4 donors and 2 recipients) he showed some very interesting results from transferring fecal microbiota (FM) from healthy individuals to those with an inflammatory bowel disease.
Among the interesting results are the fact that he was able to assemble 97 metagenomes in the 4 donor samples. Following the recipients at 4 and 8-weeks post FM transplant showed that the microbial genomes could be classed into those that transfer and colonize permissively (both recipients), those that colonize one or the other recipient, and those that fail to colonize both. Taxa alone did not explain why some microbes colonized easily, while other failed to colonize.
He also showed that 8 weeks post FM transplant, the unhealthly recipients had improved symptoms but also showed that in a PCA analysis of the composition of the recipient gut and the healthy human gut from 151 human microbiome project (HMP) samples, the recipients moved into the “healthy” HMP cluster from being extreme outliers on day 0.
He also investigated differential gene function enrichment between the permissive colonizers and the microbes that never colonized recipient’s guts and found that sporulation genes may be a negative factor driving the failure (or success) of transplantation. He proposed that the recent and notable failure of the Seres microbiome drug in clinical trials may be owing to the fact that the company killed the live cultures in favor of more stable spore-forming strains when formulating the drug. His work would suggest that these strains are less successful at colonizing new hosts.
Bo Zhang, 3D genome browser
With the ever-increasing volume of genomic and regulatory data and the complexity of that data, there is a need for accessible interfaces to it. Bo Zhang’s group at Penn State has worked to make a new type of genome browser available that focuses on the 3D structure of the genome, pulling together disparate datatypes including chromatin interaction data, ChIP-Seq, RNA-Seq, etc… You can also browse a complete view of the regulatory landscape and 3D architecture of any region of the genome. You can also check the expression of any queried gene across hundreds of tissue/cell types measured by the ENCODE consortium. On the virtual 4C page, they provide multiple methods to link distal cis-regulatory elements with their potential target genes, including virtual 4C, ChIA-PET and cross-cell-type correlation of proximal and distal DHSs.
All in all, GLBIO2017 was a very enjoyable and informative meeting where I met a lot of great colleagues and learned much. I am looking forward to next year!
Recently, I’ve been working my way through Ben Langmead’s excellent introduction to “Algorithms for DNA sequencing” on Coursera.com. The class is a fascinating and well-taught intro to concepts about DNA short read alignment and assembly methods.
As part of the course, we have implement or modify python code relating to several simple matching algorithms, including the “naive exact” (NEM) matching method, the “boyer-moore” (BM) method, and a k-mer index approach.
I was curious about speed, so I made a figure showing the computational time that each approach takes. P and T refer to the length of the short read to be aligned and the genome to align to, respectively.
Note that the y-axis is a log scale in units of microseconds. Right away, it is obvious that k-mer index methods are orders of magnitude faster than ‘online’ methods like NEM and BM.
Also of interest is the fact that as the pattern gets shorter, the advantage of BM preprocessing of the pattern gets smaller. You can see that going from 30 to 11 pattern length negates any advantage to BM searching.
I’ve had a lot of questions from researchers on how to get started with Ingenuity Pathway Analysis (IPA). I made a short video to address them: